Public Sphere Journal 2017 Edition by The Public Sphere

JOURNAL OF PUBLIC POLICY I 2017 EDITION

2017

2016-17 Editorial team Ashley Lau, Editor-in-Chief Andrea Villarreal Ojeda, Managing Editor Jonathan Melo, Lead Editor, Print Zachary Clemence, Lead Editor, Web

Senior Editors Silvana Rebaza Zachary Rosentzveig Mateo Salazar Gabriela Smarrelli Diana Zambonino

Contributing Editors ThĂŠo Bourgery Antony Declercq Xinyue Ding Joshua Eyre Felix Keser Rachel Knapp Frances MacLellan Chelsea Phipps Geoffrey Si

www.publicspherejournal.com

The Public Sphere is a journal of international policy studies produced by postgraduate students at the London School of Economics and Political Scienceâ&#x20AC;&#x2122;s Institute of Public Affairs (IPA). The Public Sphere Journal, now in its fifth year, publishes evidence-based research in economic policy, public management, international affairs, development studies and social policy. The journal offers contributors and readers an active space to rethink, critically analyse and address important policy problems. The Public Sphere welcomes submissions that contribute to vibrant debate from advanced students in relevant degree programs, academics and practitioners.

CONTENTS

Letter from the editors

A mixed-methods evaluation of Roma health mediation: Bulgaria case study Mario Battaglini

Addressing spatial segregation of the low-income population through comprehensive social housing projects in Quito, Ecuador Vanessa Alejandra Carrera Yepez

Why have UK pension funds not taken on social impact investing? Dana Elman Vishkin

Human trafficking and ethnic minority problems in Myanmar: Policy recommendations for Myanmar and neighboring states Seunghyun Han

Is the UK’s proposed soft drink tax an effective way to reduce obesity? Genevieve Joy

Are Malawi, Rwanda, Uganda and Zambia prepared to meet the Sustainable Development Goals’ health targets? A policy analysis Natasha Mbabazi

157

Equality of opportunity and fiscal policies: Where are we? Where do we need to be? David Ortan

193

The Spanish airport system: Critical evaluation of the effectiveness of the Spanish government’s management of public resources Ane Elixabete Ripoll-Zarraga

221

Monetary policy at the zero lower bound: Discretion, uncertainty and interest rate volatility Mark A.C. Wilkinson

255

Flushing, rituals and needle fixation among heroin addicts: Implications for policy John Wooldridge

279

75 107 131

[Letter from the editors]

LETTER FROM THE EDITORS Dear Reader, Policy debates are all around us; while some make big headlines, others are more subtle. The past year delivered no shortage of significant public policy discussions and debates in every corner of the world and at home, here in London. We at The Public Sphere are delighted to bring you this year’s edition of the journal, which places you at the forefront of several of these important issues – from human trafficking of ethnic minorities in Myanmar, to spatial segregation of the low-income population in Ecuador, and health policy assessment in Sub-Saharan Africa. The Public Sphere Journal, now in its fifth year, was established by postgraduate students at the London School of Economics and Political Science in the Institute of Public Affairs’ flagship Master of Public Administration programme. This year was particularly significant for the growth and innovation of the journal. We received a record number of article submissions from public policy institutions across Africa, Asia, Australia, North America, South America and Europe. We also expanded our team and built up our online presence through brand new blog series and digital content on our website that have allowed us to extend the policy conversation to a wider audience. These spaces flourished under the efforts of our team members, who believe that this journal is an ever-growing contribution to evidence-based policy through rigorous contributions to a diverse and important array of debates. The ten articles in this edition are those selected by our senior editorial team. They represent a range of topics and regions, and were the strongest in policy relevance, methodology, academic rigour and originality out of this year’s very competitive pool of submissions. In the following pages, you will be able to examine

THE PUBLIC SPHERE | 2017 ISSUE

a Roma health mediation programme in Bulgaria, assess ethnic minority problems in Myanmar, and evaluate the effectiveness of the Spanish government’s management of public resources through its airport system. Through careful examination, you will be able to explore monetary policy at the zero lower bound, dissect the equality of opportunity through fiscal policies, and think critically about spatial segregation in Ecuador’s capital city. A series of in-depth interviews with heroin users will shed important light on the development of needle fixation, while a critical analysis of health policy in Malawi, Rwanda, Uganda and Zambia will enhance your understanding of the Sustainable Development Goals in Sub-Saharan Africa. From here in the United Kingdom, you will be able to assess the effectiveness of soft drink taxation, and learn about the engagement of UK pension funds with the social impact investment market. We would like to thank our authors for their hard work and dedication, and the LSE’s Institute of Public Affairs for their support in our mission. Thank you for reading, and we hope you enjoy. Ashley, Andrea, Jonathan and Zachary

THE PUBLIC SPHERE 2017 ISSUE

A mixed-methods evaluation of Roma health mediation Bulgaria case study Mario Battaglini

MSc Social Policy Research, Class of 2016 London School of Economics and Political Science

ÂŠ Sarah Ju

ABSTRACT Efforts to bridge the health gap of Roma people compared to the general population in the EU have been renewed through the adoption of National Strategies. The Roma Health Mediator programme takes an active part in the Bulgarian Strategy by providing a range of social and health-related services. This essay evaluates the relevance, effectiveness and sustainability of such a programme. The analysis shows various interconnected strengths of the Bulgarian mediator model. It also identifies issues of fragility related to the decentralised approach focussed on municipalities, on the one hand, and to the lack of a supportive macro-level welfare system, on the other.

[A mixed-methods evaluation of Roma health mediation]

INTRODUCTION An overview of Roma health inequity in the EU offers a grim picture, as Roma lag behind in terms of mortality, morbidity, immunisation rates, health insurance coverage, and healthy lifestyles (European Commission, 2014). Efforts to tackle Roma disadvantage have been renewed through National Roma Integration Strategies, wherein the Bulgarian health mediator programme takes an active part. In this essay, I attempt to answer the evaluative questions of whether the Roma health mediator programme in Bulgaria is a relevant, sustainable and effective policy to tackle the underlying drivers of Roma disadvantage, and what the mechanisms that favour and hinder intervention are (Pawson and Tilley, 1997). In terms of relevance, I seek to ascertain whether there is a close match between the drivers of disadvantage and the activities performed, and the extent to which mediators can tailor such activities to the specific needs of their clients. As for effectiveness, I seek to provide evidence that health promotion is fostered, access to healthcare is increased, and a more fertile environment for addressing Roma needs is created. Regarding sustainability, the focus is on intercultural, organisational and financial elements. The study is divided into three parts. The first introduces the problem of Roma health inequity and its underlying drivers, which health mediators seek to address in their daily encounters with people (clients, doctors, municipal coordinators and other social workers) in the field. It also briefly considers the EU and national policy context wherein the programme takes place. The second part clarifies the reasons for adopting a mixed-methods research approach, together with a description of the data collection (survey, interviews) and analysis techniques (descriptive and inferential statistics, thematic network analysis), followed by the manner and stages at which methods were combined. The third section analyses the activities the mediators perform, with

THE PUBLIC SPHERE | 2017 ISSUE

an emphasis on what hinders and facilitates action, and what the main qualitative outcomes and quantitative outputs are. It also investigates three strongly interlocked aspects that may influence the programme, in particular the extent to which it is financially viable, well led and responsive to clients’ needs: (i) the internal and external coordination of the activities; (ii) the motivation and discretion of the mediators, which are seen as “street-level knights” to combine Lipski’s (1980) and LeGrand’s (2003) inputs; and (iii) their training and selection. Lastly, the conclusion summarises the analysis by providing recommendations. FIGURE 1 CONCEPTUAL MAP Power Money

Knowledge

Beneficial social connections

Roma health inequity

Roma Health Mediator programme

Match between drivers and activities

Relevancy

Activties tailored to clients’ needs

Financially viable

Source: Author.

Presitge

Increases access

Fosters health promotion

Effectiveness

Sustainability

Respectfully bridges tradition and law

Creates fertile environment on Roma issues

Well-led

[A mixed-methods evaluation of Roma health mediation]

BACKGROUND In Bulgaria, data indicate a strong health disadvantage for Roma populations (Gitano 2009, UNDP 2012). According to Link and Phelan’s (1995) “fundamental cause” approach; when a population develops the means to avoid disease and death, individuals’ ability to benefit from that wherewithal is “shaped by resources of knowledge, money, power, prestige and beneficial social connections”. Barriers to healthcare access are an important cause of health inequities. Such barriers could be both formal and informal, from the supply and the demand side. Firstly, regarding de jure/formal barriers, Bambra (2007) highlights how research on health inequity has increasingly integrated a welfare state perspective. Since the fall of the communist regime, Bulgaria has reformed its healthcare services vis-à-vis financing, organisation and delivery. Waters et al. (2008) show how most Central and Eastern European (CEE) countries moved away from centralised health systems and instead adopted a social health insurance model. In Bulgaria, insurance coverage was introduced in 1999, excluding the most vulnerable. Besides, in 2015, the number of months of arrears to be paid to restore one’s entitlement to insurance increased. Secondly, regarding informal barriers, the literature highlights out-of-pocket payments, lack of services or transportation to healthcare facilities, low health literacy, mistrust, discrimination, lack of accommodation of cultural differences, and Roma traditional beliefs (Atanasova et al. 2012, Colombini et al. 2012, Hajioff and McKee 2000, Zoon 2001). Such challenges have not gone unnoticed. In 2005, the Declaration of the Decade of Roma Inclusion 2005–2015 was signed. In 2011, the European Commission’s Communication on a EU Framework for National Roma Integration Strategies (NRIS) up to 2020 asked member states to devise integration strategies, to be implemented through an Action Plan. The Action Plan for the implementation of the National Roma Integration Strategy of the

THE PUBLIC SPHERE | 2017 ISSUE

Republic of Bulgaria (2012-2020) and the Decade of Roma inclusion 2005-2015 include health mediators in many activities. Mediators act as a bridge between Roma and institutions, improving access to healthcare by reaching across prejudices and overcoming lack of information. The health mediation programme was originally piloted in Kjustendil in 2001 by the Ethnic Minorities Health Problems Foundation. In 2005, with the start of the Decade, the Bulgarian government adopted the Health Strategy for Disadvantaged Persons Belonging to Ethnic Minorities, wherein the mediators played a significant role. In 2007, the mediator profession was included in the National Classification of Occupations, and supported financially by the state. In the same year, the National Network of Health Mediators (NNHM), including mediators and experts, was founded. In 2016 funds have been allocated for a total of 196 health mediators.

METHODOLOGY Mixed-methods Quantitative and qualitative approaches are traditionally understood as following different logics of enquiry due to differences in their epistemology, ontology and role of theory. This rift is at the basis of the incompatibility thesis, according to which it would be inappropriate to mix methods due to such incommensurability. However, the rejection of an ‘either-or stance’ has long been part of the mixed-methods literature (Campbell and Fiske 1959, Howe 1988), thus supporting methodological eclecticism. Mixed-methods are used for three main reasons: firstly, to confirm results through triangulation, given that “the weaknesses in each single method will be compensated by the counter-balancing strengths of another” (Jick 1979: 613). Secondly, to offer a different part of the whole, so as to add meaning and

[A mixed-methods evaluation of Roma health mediation]

understanding to the underlying mechanisms at hand. Thirdly, to clarify surprising findings; what Green, Caracelli and Graham (1989) call “initiation.” In this study, the methods were mixed throughout, rather than being compartmentalised: firstly, the literature review, documentary analysis and a cognitive interview helped to devise the survey; secondly, the results of the survey helped to sample the interviewees; and, thirdly, the quantitative findings are integrated with the qualitative ones.

Data collection The methods used are a self-completion survey administered through “qualtrics” by the mediators and 8 face-to-face semi-structured interviews with the mediators, experts of the NNHM, a representative of the Bulgarian Family Planning Association (BFPA), and with the National Council for Cooperation on Ethnic and Integration Issues (NCCEII). The main benefit of the self-completion survey is that it made available a substantial portion of the mediators with a high degree of data standardisation. Semi-structured interviews allowed getting a “worm’s eye view,” as opposed to the macro-level “bird’s eye view” made possible by survey data (Hakim 2000). In surveys, respondents have a limited ability to fully express what they mean when selecting among variables, and the fact that it is difficult to build a rapport may negatively influence the truthfulness of answers. By contrast, through probing and targeted questioning, while interviewing, the researcher obtains clarification and expansion of meanings (Gaskell, 2000). Whilst interviewing has been criticised too (Dean and Whyte 1958, Becker and Geer 1957, Deutscher 1973), answers to different questions and from different stakeholders were triangulated and individuals were interviewed in sufficient detail for the results to be taken as truthfully disclosed, correct and complete, and thus internally valid.

THE PUBLIC SPHERE | 2017 ISSUE

Regarding interview sampling, mediators from different regions, both women and men, recently employed and with previous experience, working in rural and urban settings, or who had provided a ‘deviant’ survey answer were sampled. Hence, contrary to a homogenous sample whose purpose is “to describe some particular subgroup in depth” (Patton 1990, 235), the method was a maximum variation one, which tended to identify core patterns and reasons for variation. Lastly, two interviews were conducted with experts of the National Network who introduced me to NCCEII and BFPA representatives (snowball sampling).

Data analysis The survey included two parts: the first used a 5-point Likert scale to measure the extent of agreement regarding 4 items: discretion-goal, discretion-client, mediator willingness, and job meaningfulness. The second used ranked responses vis-à-vis priority activities: improving health education, preparing documents, accompanying individuals to healthcare service centres, raising doctors’ intercultural awareness, patient advocacy towards institutions, and improving communication between Roma and medical professionals. The Likert scale items were analysed with measures of central tendency of the frequencies, highlighting the modal category. Secondly, regarding ranked responses of priority activities, descriptive statistics and a one sample z-test for evaluating proportions were carried out: improving health education was a statistically significant first choice, as discussed in the next section. Interviews were interpreted through thematic network analysis. This process entailed the classification of coding to identify global, organising and basic themes (Attride-Stirling 2001). The coding process was primarily inductive but could not be separated from a certain degree of deductive a priori templates. Relevant quotes are presented throughout this paper so as not to lose sight

[A mixed-methods evaluation of Roma health mediation]

of the rich experiences. Presentational sampling (Flick 2002, 61) was a particular challenge, as it necessitated the sacrifice of vivid descriptions for the sake of brevity.

THE ROMA HEALTH MEDIATION PROGRAMME Activities Roma health mediators are health workers hired by municipalities to advise and support Roma communities. They provide a vast range of both social and health-related services, which can be grouped into three main kinds of activities: health education, assistance in accessing healthcare and advocacy. Personal contact with a range of stakeholders is central and their work is mainly in the field â&#x20AC;&#x201C; only a small part is in the office. Mediators strive to act as a bridge between the vulnerable, poorly integrated, and lower-educated Roma communities, on the one hand, and the institutions, on the other. According to the National Network of Health Mediatorsâ&#x20AC;&#x2122; 2015 monitoring reportâ&#x20AC;&#x2122;s raw data, the total number of service instances provided was 130,657. The largest share of operations concerned assistance in carrying out prophylactic examinations, immunisations, and campaigns in the field of reproductive health and prevention of various diseases (65,931 or 50.46 per cent). The second consisted of counselling and preparation of documents, accounting for 37,500 of the outputs (28.70 per cent). The third, accompaniment to health and social services, accounted for 15,066 of the activities or 11.53 per cent. Lastly, 12,360 services (9.45 per cent) were aimed particularly at the poor, pregnant women, older people living alone, etc. At a first and rather superficial glance, the data would seem to suggest that activities in the field of accessing healthcare, rather than health education, are prioritised. However, the way the output data is collected does not make it easy to distinguish between

THE PUBLIC SPHERE | 2017 ISSUE

the two types. For instance, the counselling and documents category conflates two rather different activities. Besides, it would be incorrect to make a sharp, clear-cut division between the two: firstly, vaccinations or assistance with sexual and reproductive health programmes entail as much technical assistance as the delivery of information regarding their benefits. Secondly, by accompanying individuals to healthcare service centres, mediators not only work as interpreters and cultural brokers but also raise General Practioners’ (GP) awareness of the needs and rights of Roma, as a result creating a more fertile environment. Thirdly, by navigating Roma through the social and health services, they make them acquainted with the different systems in the process. The Network’s monitoring data, therefore, does not allow to measure whether the programme focuses on healthcare access or health education, nor can the complex nature of such activities be fully grasped by relying on these quantitative outputs. The reason why this is highlighted is because, while the group of diverse activities pertaining to health education can be seen as an investment in health, the cluster of services devoted to improving and increasing access to healthcare can be seen as favouring healthcare consumption. To an extent, it is reasonable to consider the latter activities as a somewhat inferior kind, because they are not centred on engendering long-term sustainable change, but rather on tackling short-term health needs. Yet, as discussed above, there is more to them than meets the eye. Since, based on the available monitoring data, it was hard to measure the extent to which health investment or consumption were prioritised, mediators were asked in the survey to rank six activities: improving health culture, advocacy with institutions, providing information on the health and social services, documents procurement, improving communication with GPs, and raising the awareness of medical professionals vis-à-vis Roma needs. Improving health culture stands out as the highest ranked option, as can be seen from Figures 2 and 3:

[A mixed-methods evaluation of Roma health mediation]

FIGURE 2 WORKING ACTIVITIES IN ORDER OF IMPORTANCE - 1ST CHOICE % 35.0% 30.0% 25.0% 20.0% 15.0% 10.0% 5.0% 0.0%

31.7%

19%

17.5%

Health culture

Institutions advocacy

Information

17.5%

12.7%

1.6%

Documents Communication Awareness procurement

Source: Author. FIGURE 3 WORKING ACTIVITIES IN ORDER OF IMPORTANCE - 1ST/2ND CHOICE %

50.0% 45.0% 40.0%

46%

35.0%

34.9%

30.0%

34.9% 30.2%

25.0%

27%

20.0% 15.0% 10.0% 5.0% 0.0%

Health culture

Institutions advocacy

Information

Documents Communication Awareness procurement

Source: Author.

Besides, a z-test for evaluating proportions of the same population assuming a hypothesised value of 16.6 per cent, which is equivalent to indifference among the six options, found a clearly statistically significant deviation from the indifference percentage for improving health culture as option one and as combined option one and two, significant at the 0.1 per cent and 5 per cent levels respectively. Therefore, the outputs of the monitoring activities only tell a partial story, as mediators put a lot of emphasis on empowering

THE PUBLIC SPHERE | 2017 ISSUE

the Roma – on health investment, rather than merely its consumption. Mediators not only bridge the gap between Roma and institutions, but also attempt to reduce the gap to be bridged, thus aiming at health promotion, a “process of enabling people to increase control over, and to improve, their health” (Ottawa Charter, WHO 1986). As a mediator puts it: “besides helping people, we have to inform them, so that they can help themselves eventually.” In this regard, it ought to be highlighted that, instead of accompanying people to healthcare service centres, mediators may refer the clients to the appropriate place, with the intention of having them learn: “If people need me less I think this would be a great achievement […] I tell them to go alone, do this and that, and see that nothing scary will happen, not because I don’t want to go but for them to be independent, to learn.”

According to the mediators’ self-assessment of the outcomes, lack of knowledge and of beneficial social connections (with GPs, role models, mediators) have been overcome. The outlook is not exclusively optimistic, however, as mediators emphasise that change is slow and difficult to sustain with regards to “hardly negotiable” topics because of the clash with family and community values. Evidently, the social structure plays an influential role in shaping life’s goals and health outcomes: early marriages, from 14–16 years old, are socially encouraged, and the key goal becomes having children, “so that people don’t start gossiping [that] they can’t.” However, mediators underline positive outcomes with those most eager to access new sources of information as compared to the traditional sources they would otherwise be confined to – thus, offering a chance to the autonomous agent to make a choice. The tension between individual agency and community structure and values is best described by a mediator:

[A mixed-methods evaluation of Roma health mediation]

“I understand a person is very rich when [they] know both cultures. I am a Bulgarian of Roma origin. So one is rich when [one] can choose and decide for her own children. How to raise them? Should they follow the traditional way or go beyond that in the wider society, [which is] more difficult and challenging but ultimately better?”

As for what facilitates action, firstly, it should be stressed that mediators are both men and women. This gives a fuller range of options as for who is to address specific target groups or individuals – for instance, male mediators may find it easier to talk to husbands, who ultimately take the final decisions regarding family planning and reproductive health. However, this arrangement does not hold in all contexts, as for instance only one mediator may be employed in a municipality or, if there are two or more, they may work in different neighbourhoods. Secondly, if education and job mediators are also employed in the municipality, it is possible to create synergies across the critical areas of exclusion. This is because, while mediators can be resourceful and work on many dimensions of the drivers of Roma health inequity, they cannot tackle each of them, such as material causes linked to entrenched poverty and lack of insurance: “When a woman needs to change the spiral, I use my personal connections so that I can get an appointment with the doctor, but the woman still has to pay 15 leva. There needs to be an examination before inserting [the pap smear], and the examination is 20 leva. But they do not have the 20 leva for it, so they do not put the spiral sometimes.”

As for lack of insurance, mediators emphasise that emergency treatment (in accordance with Decree 17 of 31st December 2007) and one visit to pregnant women (in accordance with Ordinance 26 of 14th June 2007) are accessed primarily through their sup-

THE PUBLIC SPHERE â&#x20AC;&#x2020;| â&#x20AC;&#x2020;2017 ISSUE

port. Roma clients would hardly know about such entitlements otherwise. The third facilitating mechanism is linked to the constant work with certain GPs. This has improved the relationship between physicians and the mediators, as well as between GPs and Roma. Mediators were keen to emphasise that the most troubling relationships are with medical specialists, rather than with GPs. The improved awareness and intercultural skills of GPs is a result of sustained cooperation. Similarly, social services appear to have become more responsive, thus decreasing meso-level barriers to accessing welfare. An opportunity to further decrease these barriers stems from the fact that health mediators in certain municipalities have started to work in local hospitals, too.

Coordination The financial and activity-related coordination of the Roma health mediator programme takes place at different levels. The programme has sustainable features in that the mediators are paid by the Ministry of Health through the Ministry of Financeâ&#x20AC;&#x2122;s delegated budget. However, the budget is approved on a yearly basis, rather than on a multi-annual one. Despite there being room for improvements in this regard, so as to plan in the medium to long term, the current financing structure marks an improvement from the period of the early 2000s, when ad-hoc projects were the channel through which to train and hire mediators. Nonetheless, the National Network of Health Mediators, together with its partner organisations, constantly look for further sources of financing from external donors, which could be international, foreign, as well as Bulgarian, and public as well as private. The mediators, through the abovementioned delegated budget, are employed by the municipalities. Each year the National Network of Health Mediators prepares a list with the municipalities that should receive financing by the state, based on

[A mixed-methods evaluation of Roma health mediation]

their needs-analysis and monitoring activities. This list is sent to the Ministry of Finance and the Ministry of Health. A great deal of coordination is possible thanks to the work of the National Council for Cooperation on Ethnic and Integration Issues (NCCEII). The NCCEII is the consultative and coordinative body of the Council of Ministers. It helps to implement and monitor the National Roma Integration Strategy and its Action Plan. Furthermore, there exist 28 regional strategies (one per region) and more than 300 municipal action plans (out of a total of 365 municipalities). “The regional strategies and municipal action plans follow the structure of the National Strategy, provide information on local Roma communities, and specify the national aims depending on the different contexts” (IOM 2014, 19). The National Network of Health Mediators is accountable to the NCCEII, to which it sends a report every year. The mediators send a report every six months to the National Network; it is on the basis of these reports that the final yearly report to be sent to the NCCEII is compiled. Coordination may also take place rather informally: “Members of the National Council ought to submit a short yearly report of their activities for monitoring purposes. But when we need some special information we can get in touch more directly.” – NCCEII

In addition, mediators send a monthly report to the municipality that employs them. The extent to which the various municipalities are eager to support the work of mediators, or are interested in and knowledgeable about their activities and issues at stake, changes widely across the country and over time. There is also a degree of variability in the extent to which regions operate proactively:

THE PUBLIC SPHERE | 2017 ISSUE

“It’s a bit of everything, there are regions that are very active and where mediators are included in different regional initiatives, funded by different donors, there are regions in which things are more quiet” - BFPA

Also, regarding health planning at the local level, the views of the mediators pertaining to their inclusion vary: “When they write the strategy they would ask me to sit together to support, inviting me in working groups so that my opinion is heard and taken into account.” “So far they have kept us on the side. In three or four years, they have only invited me to one meeting.”

Taking a bird’s eye view of the process of coordination and strategic planning of activities, it seems that the way priorities are set takes into account the interplay of different actors and the context wherein the planning process takes place (Buse 2005), thus shying away from a rigid top-down, rational model of strategic planning of activities (Simon 1945, Marinetto 1998). The internal coordination of the Network favours such a worthwhile approach. The Network has an online forum in place, which allows the mediators to exchange information quickly and horizontally, rather than only vertically (mediator-Network) every 6 months. Besides, Network coordinators perform field visits to different geographical areas throughout the year. Therefore, the Network is in a position to inform policy at the national level by having close links with the local mediators – and awareness of the needs of the Roma – and by having both formal and informal linkages with the National Council. Moreover, mediators are richly resourceful in the information they may provide regarding the structural, cultural, and health needs of the Roma. The job description requires the mediators

[A mixed-methods evaluation of Roma health mediation]

to be from the community they work in. Therefore, even if the mediator has just been employed, she or he possesses a personal experience of the context. Further to this, the fact that the job description requires them to have at least a high school education allows them to have a critical understanding of the problems, enhanced by the thorough training they receive. In the municipalities where mediators have long been active, more experienced mediators help new recruits to better understand the context, as they possess a longitudinal experiential background related to the community. This bottom-up process is best expressed in the words of a mediator: “We know the territory and the people, we know who is sick, what needs there are in the community, and that’s how we come to know: when we go there and see and talk to people.”

Externally, the National Network of Health Mediators has created a network of partnerships with other NGOs. One of the NGOs that most contributed to the activities of the Network is the BFPA, the national affiliate of the International Planned Parenthood Federation (IPPF), which has been active in Bulgaria for 25 years. Such a policy community resonates well with Lewis’s (2005) idea of it, inasmuch as it has developed through formal and informal interaction of its participants across time.

Street-level knights Social and public policies are best understood in the daily encounters of street-level workers (Lipsky 1980) rather than as a top-down implementation of laws by high ranking administrators, albeit with varying degrees of flexibility or participatory components. This is because front-line staff interact directly with citizens, have discretion in their job, and thus are active makers of policy. Hence, the way the policy under investigation is ultimate-

THE PUBLIC SPHERE | 2017 ISSUE

ly shaped on the field depends heavily on whether the mediators are motivated (have “willingness”) in their job, and what affects their motivation. It has been argued that street-level bureaucrats want to make a difference to their clients’ lives (Maynard-Moody and Musheno, 2000), thus portraying their job as “meaningful”. This in turn increases their motivation to implement the policy (Tummers and Bekkers 2014). With regard to discretion, the survey separated the concept between treatment of the client (discretion-client ) and discretion towards the goal to be achieved (discretion-goal ). The mediators appear to perceive a very strong degree of discretion. In both instances the modal category was ‘strongly agree’ (60.32 per cent for discretion-client, and 63.49 per cent for discretion-goal), thus supporting the literature’s insight that front-line staff generally enjoy a high degree of discretion. Conversely, a mediator who felt she could not tailor her job to client needs (discretion-client) explained in the subsequent in-depth interview that: (i) the high number of clients did not allow for adapting the intervention to the needs of each client; (ii) it would have been preferable to have the office in the Roma quarter so as to be closer to the community and thus more effective in performing the duties; and (iii) there had been instances where the needs of certain clients (drug addicts, a mother whose child had died in an institution), who had turned particularly aggressive, could not be addressed. This powerful explanation bears important policy implications. Firstly, it highlights that the programme is understaffed in some contexts. Secondly, it underlines the tension between creating an ad-hoc Roma structure and favouring mainstream healthcare access. Thirdly, it supports the understanding that mediators face dire challenges in the job, including violence and threat of violence, and thus require strong commitment to overcome them. Indeed, for “meaningfulness” and “willingness” , the modal category is ‘strongly agree’ (58.7 per cent and 74.60 per cent), with cumulative percentages of ‘strongly agree’ and ‘agree’ of 98.4

[A mixed-methods evaluation of Roma health mediation]

per cent and 93.7 per cent, respectively. According to the mediators, their profession is more of a vocation than a job. The fact that the mediators are Roma themselves goes a long way in explaining such dedication: “I have accepted it as a mission. I didn’t start for the money, I didn’t need any, my husband has his business, the salary is very bad.” “One of the biggest challenges for the Roma community is health. So, I wanted to be of help to my people […] I thought I could be a saviour.”

Mediators can be primarily seen as public-spirited altruists – as knights rather than knaves, to use LeGrand’s (2003) distinction vis-à-vis healthcare service employees. According to LeGrand, “knaves can be defined as self‐interested individuals who are motivated to help others only if by so doing they will serve their private interests; whereas knights are individuals who are motivated to help others for no private reward, and indeed who may undertake such activities to the detriment of their own private interests” (LeGrand 2003, 27). Indeed, the programme tends to select knights and dissuades knaves (low pay, heavy workload, threats of violence, non-monetary rewards, etc). However, knights and knaves coexist within any organisation (ib. 2003). For some mediators the job is merely a temporary activity while studying or upon entering the job market. In this case, the same aspects that select for knights prove counterproductive. This is because the challenges faced negatively impact staff turnover, thus squandering the resources used for recruitment and training, and those forged with clients and other stakeholders, jeopardising the sustainability of the programme. What is even more problematic is that such weakness is not limited to temporary job-seekers alone. As the interviews clarified, turnover due to the programme being understaffed and the workload

THE PUBLIC SPHERE | 2017 ISSUE

too burdensome may as well affect knightly mediators. Digging deeper, as highlighted by the representative of the BFPA, “there are also places with lower numbers of Roma per mediator, but worse problems” – thus, the quantitative approach to caseload only tells a partial story, but cannot be overlooked altogether in shedding light on turnover or effectiveness.

Selection and training The criteria for the selection of health mediators include being from the community they work in, a minimum of high school education, and the ability to speak the language of the local community. Every year, the Network provides the relevant Ministries with the number of mediators needed. A budget is allocated by the Ministry of Finance and received by the municipalities. A municipal commission is set up to examine the candidates for the job openings, which include representatives of the municipality, the regional health inspectorate and mediators. However, certain municipalities do not want to include the mediators in such commissions. According to various interviewees, it is very difficult to work with municipalities because local political motives may prevail. By contrast, it is easier to work at the national level, as it was at the beginning of the programme, before the municipalities were put in charge. Therefore, while there is a structured coordination system, there seem to be threats of fragility. Similarly, critical issues pertaining to the firing of mediators persist. In one municipality, the mayor wanted to remove a very proactive and capable mediator with nine years of experience and substitute them with an individual who had supported him in the electoral campaign, according to a representative of the National Network. Conversely, there are also instances where less skilful mediators, according to the monitoring activities of the Network, are kept in place because their inability to bring about change matches the (lack of) will or enthusiasm of the political elites. The

[A mixed-methods evaluation of Roma health mediation]

National Network has to intervene on a case by case basis in these circumstances, so as to defend the rights of the mediators and the effectiveness of the overall programme. In such cases, the Health Commission of Parliament, the National Council, the Ministry of Finance, and the Ministry of Health are informed in order for them to take action. It would be advisable to put in place a strengthened hybrid system, wherein mediators who perform above a given threshold, according to the monitoring activities of the Network, cannot be fired on a legal basis. Nonetheless, such proposals fall short of countering another threat as, in some cases, the municipalities have insisted that mediators sign a resignation letter while being (falsely) promised another job. The Network has now thoroughly counteracted such practices by informing the mediators not to sign any documents without first informing the Network. An interviewee referred to the Belgian experience as a good practice for comprehensively countering such threats. In Belgium, the NGO Foyer is independent from the local authorities, and the whole process of hiring, retaining or firing, and assigning activities to the mediators, is more transparent, not linked to political goals, and therefore more effective. Despite this, issues pertaining to path dependency may arise â&#x20AC;&#x201C; it could be very costly at this point to change the overall structure altogether. Besides, the fact that the municipalities are involved may also present positive aspects, as this expresses (or does not) the local political will, on the one hand, and because it may favour the education of the political elites given such institutionalization, on the other. Taking all this into consideration, a hybrid, somewhat consociational structure, between the Network and the municipalities, could represent a second-best option, and an improvement compared to the current threats and the exhausting need to act on a case by case basis. A more thorough comparison of the Belgian and Bulgarian experiences, as a â&#x20AC;&#x153;most different systemâ&#x20AC;? design (Przeworski&Teune 1970), could be worthwhile.

THE PUBLIC SPHERE | 2017 ISSUE

The training of the mediators seems apt to increase their professionalism and capability. Training consists of classes in a medical university, it is 2-weeks long, and includes several hours of medical training with professionals. This is probably the most striking difference, and strength, of the training as it is now compared to how it used to be. While there has always been training for the mediators, in recent years it has included a more thorough medical education. The way the training is now organised is multidimensional: health-related information is coupled with a more practical component, such as information about the health system, health insurance legislation, social security, communication skills, how to report, how to organise time and activities, etc. Indeed, the role of mediators presupposes and requires such a synergetic mix. The training is prepared by the Network, as a mediator puts it: “The Network is a source of information, we could not manage without them, we could not work without them […] We don’t lack training, our colleagues prepare training for us very diligently.”

Crucially, there are also ample opportunities for continuous training updates while being employed, both through the Network or the BFPA, which includes the mediators in their trainings focussed on prevention of cervical cancer, or hepatitis B and C, among others, or the Council of Europe’s ROMED training, in which mediators participate on a rotating basis. By contrast, a weakness of the training is that it only addresses the mediators. With the initial PHARE programme, wherein the programme was piloted, service providers were also included.

[A mixed-methods evaluation of Roma health mediation]

CONCLUSION Roma health mediators possess a warm and communicative working culture, which they put to use to mediate conflict, misunderstanding and distrust; to protect the rights of their clients; and to enhance their clients’ knowledge of, and beneficial connections with, the health system and social services. In so doing, they provide an answer to the micro- (community, individual) and meso-level (institutional) drivers of the “unequal distribution of opportunities to lead a flourishing life” (WHO 2008) – from both the supply and demand sides, and in a bottom-up yet institutionalised fashion. The programme is not a panacea, however, as certain challenges require time to be effectively countered or, like poverty, fall outside its scope. Nor can mediators do more than cultivating health promotion or help bridge the gulf between de jure entitlements and de facto access. Therefore, the macro-level context ought to be consistent with the objective of Roma inclusion. By contrast, insurance policy, with the increased amount of arrears required to be paid to restore one’s health insurance, clashes with the goals set in the National Roma Integration Strategy. It is important to stress the risk of merely opening a “can of health mediators” in order to solve Roma problems, although the mediators do indeed perform an extremely important job – a number of jobs, in fact. There can be good and bad mediators. Mediators’ motivation and perceived job meaningfulness for their community of belonging is crucial in order to overcome the material and emotional hardships of the profession. Hence, it is important that a fair, meritocratic, and apolitical system to hire and retain mediators be put in place: (i) by redistributing powers to the Network to the detriment of the municipality in a more consociational structure; (ii) by hiring mediators and taking into account the number of clients and the pervasiveness of the contextual problems; (iii) by paying higher salaries, to retain well qualified and experienced

THE PUBLIC SPHERE | 2017 ISSUE

mediators. Furthermore, working on a longer strategic horizon is required for sustainability, and multi-annual financing should replace the current, year-by-year framework. By intervening in their own community, mediators reduce the risk of a “white man’s burden” approach, which forcefully simplifies contextual complexities and splits into a Manichean light and darkness. By contrast, mediators are a bridge between tradition and law. Their approach is a dedicated and delicate one to spanning the “us versus them” gap: understanding where it is possible to act directly, and what the “hardly negotiable” topics to be dealt with more diplomatically are. Their voices are particularly valuable when heard and heeded throughout the health planning stages. Mediators’ inputs are rich and insightful because of their own background and training. Planning more broadly, however, should seek to include a wider range of stakeholders, so as to create a fertile environment that counters anti-gypsyism. More mediators could be employed in hospitals, so as to improve the difficult relationship with medical specialists, similarly to what has been done with GPs who constantly work with mediators and Roma patients. In conclusion, the programme is highly relevant because it allows practitioners to tailor interventions to the needs of their clients, and to tackle four out of the five main drivers of health inequity, namely: lack of knowledge (health promotion, information), beneficial social connections (with service providers, Roma role models), power (advocacy, policy community) and prestige (sustained cooperation, decoupling of stereotypes). It cannot tackle households’ lack of capital. It is highly sustainable culturally because it strives to bridge tradition and law without creating a parallel system and without being biased by a “white man’s burden” approach; it is sustainable financially, but a multi-annual budget would be an improvement; and it is only somewhat sustainable in leadership, because of the threat of municipalities’ uncooperativeness. The programme is highly effective in increasing

[A mixed-methods evaluation of Roma health mediation]

healthcare access, and effective in engendering long-term health promotion. It is somewhat effective in creating a fertile environment, as the scope of training needs to be broadened.

REFERENCES Agresti, A and Finlay, B. (2014) Statistical Methods for the Social Sciences. 4th ed. Upper Saddle River, NJ: Prentice Hall. Abrahamson, P. (1999) The welfare modelling business. Social Policy & Administration, 33 (4), 394-415. Atanasova, E., Pavlova, M., Moutafova, E., Rechel, B., and Groot, W. (2013) Out-of-pocket payments for health care services in Bulgaria: financial burden and barrier to access. The European Journal of Public Health, 23 (6), 916-922. Attride-Stirling, J. (2001) Thematic Networks: An Analytic Tool for Qualitative Research. Qualitative Research, 1 (3), 385-405. Bambra, C. (2007) Going beyond The three worlds of welfare capitalism: regime theory and public health research. Journal of epidemiology and community health, 61 (12), 1098-1102 Becker, H., and Geer, B. (1957) Participant observation and interviewing: A comparison. Human organization, 16 (3), 28-32. Bernard, H. R. (2011) Research methods in anthropology: Qualitative and quantitative approaches. 5th ed. Lanham, Md: Rowman Altamira. Betancourt, J. R., Green, A. R., Carrillo, J. E., & Park, E. R. (2005) Cultural competence and health care disparities: key perspectives and trends. Health affairs, 24 (2), 499-505. Biesta, G. (2010) Pragmatism and the philosophical foundations of mixed methods research. Sage handbook of mixed methods in social and behavioral research, 2, 95-118. Bryman, A. (2003) Research methods and organization studies. 20, Routledge. Bryman, A. (2006) Integrating quantitative and qualitative research: how is it done? Qualitative research, 6 (1), 97-113. Bryman, A. (2008) Social research methods. Oxford university press. Broadhead, R. S., and Rist, R. C. (1976) Gatekeepers and the social control of social research. Social Problems, 23 (3), 325-336. Burgess, R. G. (1991) Sponsors, gatekeepers, members, and friends: Access in Educational Settings. In: W. B. Shaffir and R. A. Stebbins, (eds.) Experiencing fieldwork: An inside View of Qualitative Research, Newbury Park, CA: Sage, 43â&#x20AC;&#x201C;52. CAHROM. (2016) The ad-hoc Committee of experts on Roma issues, Thematic report of the group of experts on Roma health mediators, Strasbourg. Campbell, D. T., and Fiske, D. W. (1959). Convergent and discriminant validation by the multitrait-multimethod matrix. Psychological bulletin, 56 (2), 81-105. Colombini, M., Rechel, B., and Mayhew, S. H. (2012). Access of Roma to sexual and reproductive health services: qualitative findings from Albania, Bulgaria and Macedonia. Global public health, 7 (5), 522-534. Corsi, M. et al. (2008) Ethnic minority and Roma women in Europe: a case for gender equality? Luxembourg: Publications Office of the European Union.

THE PUBLIC SPHERE | 2017 ISSUE

Council of Europe (CoE) database of policies and good practices. Available from: http:// goodpracticeroma.ppa.coe.int/en/good-practice/health-mediator-bulgaria Council of Ministers of Republic of Bulgaria. (2005) National Health Strategy for Disadvantaged Persons Belonging to Ethnic Minorities. Sofia, Bulgaria. Creswell, J. W., Klassen, A. C., Plano Clark, V. L., and Smith, K. C. (2011) Best practices for mixed methods research in the health sciences. Bethesda (Maryland): National Institutes of Health, 2094-2103. CSDH, WHO. (2008) Closing the Gap in a Generation: Health Equity Through Action on the Social Determinants of Health. Final Report of the Commission on Social Determinants of Health. Geneva: WHO. Dean, J., and Whyte, W. (1958) How do you know if the informant is telling the truth? Human Organization, 17 (2), 34-38. Rorke, B et al. (2015) A Lost Decade? Reflections on Roma Inclusion 2005-2015. Decade of Roma Inclusion Secretariat Foundation. Denscombe, M. (2008) Communities of Practice: A Research Paradigm for the Mixed Methods Approach. Journal of mixed methods research, 2(3), 270-283. Deutscher, I. (1973) What We Say/What We Do: Sentiments & Acts. Glenview, Illinois: Scott, Foresman & Co. Esping-Andersen, G. (1990) The Three Worlds of Welfare Capitalism. Cambridge: Polity Press. European Commission. (2010) The social and economic integration of the Roma in Europe. European Commission. (2011). An EU Framework for National Roma Integration Strategies up to 2020. European Commission. (2012). National Roma Integration Strategies: a first step in the implementation of the EU Framework. European Commission. (2013). Steps Forward in Implementing National Roma Integration Strategies. European Commission. (2014). Report on the health status of the Roma population in the EU and monitoring data collection in the area of Roma health in the Member States. European Monitoring Centre on Racism and Xenophobia (EUMC). (2003) Breaking the barriers—Romani women and access to public health care. Luxembourg: Office for Official Publications of the European Communities. Flick, U. (2009) An introduction to qualitative research. 4th ed. Sage Publications. Fowler, F. J. (1993) Survey research methods. 2nd ed. Newbury Park, CA, London, and New Delhi: Sage Publications. Gaskell, G. (2000) Individual and group interviewing. In: M. Bauer & G. Gaskell (eds) Qualitative Researching with Text, Image and Sound: A Practical Handbook for Social Research. London: Sage Publications, 38–56. Gitano, Fundacion Secretariado (eds.) (2009) Health and the Roma community, analysis of the situation in Europe. Bulgaria, Czech republic, Greece, Portugal, Romania, Slovakia, Spain. Cuadernos técnicos, (97). Goertz, G. and Mahoney, J. (2012) A tale of two cultures: Qualitative and quantitative research in the social sciences. Princeton University Press. Green, J. C., Caracelli, V. J., and Graham, W. F. (1989) Towards a conceptual Framework for Mixed-Method Evaluation Designs. Educational Evaluation and Policy Analysis, 11 (3), 255-274.

[A mixed-methods evaluation of Roma health mediation]

Greene, J. C., and Caracelli, V. J. (1997a) Defining and describing the paradigm issue in mixed-method analysis. In J. C. Greene & V. J. Caracelli (eds.) Advances in mixed-method evaluation: The challenges and benefits of integrating diverse paradigms. New Directions for Evaluation (74), 5-17. San Francisco: Jossey-Bass. Greene, J. C., and Caracelli, V. J. (1997b) Crafting mixed-method evaluation designs. New directions for evaluation, (74), 19-32. Groves, R. M.pf, C. et al. (2009) Survey methodology. 2nd ed. John Wiley & Sons. Hajioff, S. and McKee, M. (2000) The Health of the Roma People: A Review Of The Published Literature. Journal of Epidemiology and Community Health, 54 (11), 864–869. Hakim, C. (2000) Research design: successful designs for social and economic research. Psychology Press. Howe, K. R. (1988) Against the quantitative-qualitative incompatibility thesis or dogmas die hard. Educational researcher, 17 (8), 10-16. Hopf, C. (2004) Qualitative interviews: An overview. In: Flick,U. Kardoff, E.V and Steinke I. (eds.). A Companion to Qualitative Research, 203-208. IOM Regional Office Brussels, Migration Health Division. (2014) Implementation of the National Roma Integration Strategy and Other National Commitments in the Field of Health. Bulgaria: A multi-stakeholder perspective report on 2005–2014 developments. Available from: http://equi-health.eea.iom.int/images/NRIS_Bulgaria_final. pdf Jick, T. D. (1979) Mixing qualitative and quantitative methods: Triangulation in action. Administrative science quarterly, 24 (4), 602-611. Johnson, R. B., Onwuegbuzie, A. J., and Turner, L. A. (2007). Toward a definition of mixed methods research. Journal of mixed methods research,1 (2), 112-133. JSI-Romania Research and Training Institute; Center for Health Policies and Services. (2006) Research-evaluation of the health mediation program within Roma communities in Romania. Bucharest. Available from: http://romania.jsi.com/Docs/ eval_hm_program.pdf Kelly, J.A., Amirkhanian, Y.A., Kabakchieva, E., Csepe, P., Seal, D.W., Antonova, R., Mihalov, A. and Gyukits, G. (2004) Gender roles and HIV sexual risk vulnerability of Roma (Gypsies) men and women in Bulgaria and Hungary: an ethnographic study. AIDS Care: Psychological and Socio-medical Aspects of AIDS/HIV, 16 (2), 231–245. Krumova, T. and Ilieva, M. (2008) The health status of Romani women in Bulgaria, Veliko Turnovo. Center for Interethnic Dialogue and Tolerance “Amalipe”. Available from: http://amalipe.com/files/publications/Health%20report_AMALIPE.pdf Kühlbrandt, C., Footman, K., Rechel, B., and McKee, M. (2014) An examination of Roma health insurance status in Central and Eastern Europe. European Journal of Public Health, 24 (5), 707-712. Le Grand, J. (2006) Motivation, Agency, and Public Policy: of Knights and Knaves, Pawns and Queens. Oxford University Press, U.S.A. Lewis, J.M. (2005) Health Policy and Politics: Networks, Ideas and Power. Melbourne: IP Communications. Lim, T. A., Marinova, L., Kojouharova, M., Tsolova, S., and Semenza, J. C. (2013) Measles outbreak in Bulgaria: poor maternal educational attainment as a risk factor for medical complications. The European Journal of Public Health, 23 (4), 663-669. Link, B. G., and Phelan, J. (1995) Social conditions as fundamental causes of disease. Journal of Health and Social Behaviour, 80-94. Lipsky, M. (1980) Street Level Bureaucracy. New York: Russell Sage Foundation.

THE PUBLIC SPHERE | 2017 ISSUE

Maynard-Moody, S., & Musheno, M. (2000) State agent or citizen agent: Two narratives of discretion. Journal of Public Administration Research and Theory, 10 (2), 329-358. Matrix Knowledge (2014) Roma Health Report. Available from: http://ec.europa.eu/ health/social_determinants/docs/2014_roma_health_report_en.pdf Masseria, C., Mladovsky, P., and Hernández-Quevedo, C. (2010) The socio-economic determinants of the health status of Roma in comparison with non-Roma in Bulgaria, Hungary and Romania. The European Journal of Public Health, 20 (5), 549-554. McEwen, B. S. (1998) Stress, adaptation, and disease: Allostasis and allostatic load. Annals of the New York Academy of Sciences, 840 (1), 33-44. Morgan, D. L. (2007) Paradigms lost and pragmatism regained methodological implications of combining qualitative and quantitative methods. Journal of Mixed Methods Research, 1 (1), 48-76. Niglas, K. (2010) The multidimensional model of research methodology: an integrated set of continua. Handbook of Mixed Methods in Social and Behavioural Research, 215-236. Open Society Foundation. (2005) Mediating Romani Health 2005. Available from: https:// www.opensocietyfoundations.org/sites/default/files/roma_health_mediators.pdf Open Society Foundation. (2011). Roma Health Mediators: Successes and Challenges. Available from: https://www.opensocietyfoundations.org/sites/default/files/roma-health-mediators-20111022.pdf Patton, M. Q. (1990) Qualitative evaluation and research methods. Sage Publications. Pawson, R., and Tilley, N. (1997) Realistic evaluation. Sage Publications. Powell, M., and Barrientos, A. (2011) An audit of the welfare modelling business. Social Policy & Administration, 45 (1), 69-84. Przeworski, A., and Teune, H. (1970) The logic of comparative social inquiry. John Wiley & Sons. Rechel, B., Blackburn, C., Spencer, N.J. (2009). Access to health care for Roma children in Central and Eastern Europe: findings from a qualitative study in Bulgaria. International Journal for Equity in Health, 8 (1), 24. Rechel B, Spencer N, Blackburn C, Holland R, Rechel B. (2009) Impact Of Health Reforms on Child Health Services In Europe: The Case of Bulgaria. European Journal for Public Health, 19, 326–330. Rice, M. F. (2007) A post-modern cultural competency framework for public administration and public service delivery. International Journal of Public Sector Management, 20 (7), 622-637. Ringold, D., Orenstein, M. A., and Wilkens, E. (2005) Roma in an expanding Europe: breaking the poverty cycle. World Bank Publications. Roma Decade Progress Report Bulgaria (2013) Available from: http://www.romadecade. org/cms/upload/file/9762_file2_bg-2013.pdf Robson, C. (2011) Real World Research: A resource for Users of Social Research Methods in applied settings. 3rd ed. John Wiley & Sons. Sale, J. E., Lohfeld, L. H., and Brazil, K. (2002) Revisiting the quantitative-qualitative debate: Implications for mixed-methods research. Quality and Quantity, 36 (1), 43-53. Scheppers, E., Van Dongen, E., Dekker, J., Geertzen, J., and Dekker, J. (2006) Potential barriers to the use of health services among ethnic minorities: a review. Family Practice, 23 (3), 325-348.

[A mixed-methods evaluation of Roma health mediation]

Tarnovschi, D., Preoteasa, A. M., Pamporov, A., Kabakchieva, P., & Gitano, F. S. (2012) Roma from Romania, Bulgaria, Italy and Spain between Social Inclusion and Migration. Eu Inclusive. Available from: http://www.romeurope.org/IMG/pdf/Migration_fin.pdf Tomova, I, and Nikolova, C. (2011). In: The mirror of difference: health status of the Roma and access to healthcare. Bulgarian Academy of Sciences â&#x20AC;&#x201C; Institute of Population and Human studies. Tourangeau, R., Rips, L. J., & Rasinski, K. (2000) The psychology of survey response. Cambridge University Press. Tummers, L., & Bekkers, V. (2014) Policy implementation, street-level bureaucracy, and the importance of discretion. Public Management Review, 16 (4), 527-547. UNDP (2002). Avoiding the Dependency Trap. The Roma in Central and Eastern Europe. Available from: http://hdr.undp.org/sites/default/files/avoiding_the_dependency_trap_en .pdf UNDP, European Union Agency for Fundamental Rights. (2012) The Situation of Roma in 11 EU Member States: Survey Results at a Glance. Luxembourg: Publications Office of the European Union. Verrept, H. (2008) Intercultural Mediation at Belgian Hospitals. In GarcĂŠs, V. Martin, C. and A. (eds.) Crossing Borders in Community Interpreting: Definitions and Dilemmas. Amsterdam: John Benjamins, pp. 187-201. Waters, H. R et al. (2008) Health insurance coverage in Central and Eastern Europe: trends and challenges. Health Affairs, 27(2), 478-486. WHO Commission on Social Determinants of Health. (2008) Closing the Gap in a Generation: Health Equity Through Action on the Social Determinants of Health: Commission on Social Determinants of Health Final Report. World Health Organization. WHO. (2010) How health systems can address health inequities linked to migration and ethnicity. World Health Organization: Copenhagen, Denmark. Zoon, I. (2001) On the margins. Roma and public services in Romania, Bulgaria and Macedonia. New York: Open Society Institute.

Addressing spatial segregation of the lowincome population through comprehensive social housing projects in Quito, Ecuador Vanessa Alejandra Carrera Yepez

Master of Public Administration in International Development, Class of 2016 London School of Economics and Political Science

ÂŠ Cristian Ibarra Santillan

Photo modified to black and white

ABSTRACT Since the 1970s, urban sprawl in Quito has been accompanied by the valorisation of urban land in central areas, which has led to the spatial segregation of low-income populations towards peripheral areas of the city. The present policy paper is aimed at analysing the main consequences of spatial segregation, as well as the public policies required to address them. The method used draws on secondary analysis, including both qualitative and quantitative sources. Using the theory of change as an analytical framework, the research presented identifies a causal chain that links the specific goals regarding social inclusion in Quito, Ecuador, with the conditions and interventions needed to accomplish them. Spatial segregation has several consequences in economic, social, and psychological dimensions of low-income inhabitantsâ&#x20AC;&#x2122; lives. These consequences include limited job opportunities, low-quality education, reduction of their labour supply and income, among other constraints they face to overcome poverty. Additionally, spatial segregation intensifies social problems such as crime, violence, and addiction. This phenomenon also affects the poor psychologically since it reduces their self-confidence and limits their aspirations. The analysis of theoretical and empirical evidence regarding policy options leads to the conclusion that the optimal intervention to address spatial segregation is the implementation of comprehensive social housing projects in the centre of the city. To ensure the success of these projects, their design should incorporate the establishment of mixed-income neighbourhoods, mechanisms to promote social mobility and reduce dependency, minimum quality standards for decent dwellings, and programs that guarantee the affordability of housing for the poorest tenants. Finally, these projects must be complemented by inclusive employment programs aimed at increasing information, skills, and job options for social tenants.

[Addressing spatial segregation in Quito, Ecuador]

INTRODUCTION The “right to the city” is the right of a city’s inhabitants not to be excluded from its centrality and movement (Lefebvre, 1970:175). This implies the right to inhabit, work, and freely use time and places in the city. However, urbanization patterns in contemporary cities such as Quito, Ecuador, have led to spatial segregation of the low-income population towards peripheral areas, preventing them from exercising their right to the city. This paper seeks to identify the primary consequences of spatial segregation for the disadvantaged and the public policies necessary to address them, focusing specifically on Quito. The first objective of this research is to recognise the main social, economic, and psychological effects of spatial segregation on the lives of the poor. The second objective is to identify the interventions needed to address these consequences and integrate the segregated population into the city’s urban dynamics. Quito has 2,597,989 inhabitants, 30 per cent of which are considered poor (INEC, 2010a). Patterns of urban development in the city have created obstacles for the access of disadvantaged inhabitants to housing and labour markets in the city centre. Urban sprawl in Quito has been accompanied by increasing valorisation of land in central areas due to the concentration of employment, commercial activities, and high-quality services and infrastructure. Conversely, land in peripheral areas is sold at low prices due to its remoteness from employment centres and low provision of services. Consequently, most low-income households have been forced to live in peripheral areas. Spatial segregation has several consequences for the development of low-income people. First, their remoteness from employment centres reduces job opportunities. Additionally, low-income students in peripheral areas have limited access to high-quality education. The lengthy commute made by segregated people on a daily basis reduces the amount of time they

THE PUBLIC SPHERE â&#x20AC;&#x2020;| â&#x20AC;&#x2020;2017 ISSUE

could spend on productive activities to overcome poverty. Spatial segregation also leads the poor to establish informal settlements, which generates land tenure insecurity and further obstructs their access to services, infrastructure, and employment. Another consequence is the concentration of social problems such as crime, violence, and addiction in peripheral neighbourhoods. Finally, isolation from those with different social backgrounds leads the poor to limit their educational and employment aspirations. The aforementioned problems demand urgent and comprehensive solutions from local policy makers. To identify these solutions, the present document begins by analysing the patterns of spatial segregation and its consequences in greater detail. Subsequently, I study different policy interventions such as the provision of infrastructure and services in peripheral areas, the regularization and relocation of informal settlements, the provision of social housing in central areas, and the implementation of inclusive employment policies. The advantages and disadvantages of such policies are assessed under the framework of growth management, with the objective of constructing an inclusive, sustainable, and compact city. The study concludes by recommending the implementation of comprehensive social housing projects in central areas of the city, aimed not only at correcting inequalities in the housing sector but also at integrating the low-income population into the labour market.

METHOD The research question is addressed through a literature-based analysis of theory and evidence regarding the main consequences of spatial segregation of the low-income population and the optimal policy response. The literature-based approach is complemented with the analysis and interpretation of statistical in-

[Addressing spatial segregation in Quito, Ecuador]

formation from the National Census of Population and Housing of Ecuador and data provided by Quito’s local government. This quantitative information is analysed spatially using geographical information systems, which allows us to contrast the theory with geographical patterns emerging in Quito. The analytical framework for this research is the “theory of change”, understood as “an articulation of how and why a given intervention will lead to specific change” (Stein and Valters, 2012:2). Specifically, research presented here articulates how and why the public policies recommended will address consequences of spatial segregation. To this end, I follow the recommendations of The Centre for Theory of Change (2013) to identify the desired goals regarding social inclusion in Ecuador’s capital city. Subsequently, I identify a causal chain that links these goals with the conditions and interventions needed to accomplish them. The key variable studied in this research is the poverty rate, measured using the method of Unsatisfied Basic Needs. This variable is mapped at the district level to illustrate the issue of spatial segregation of the low-income population in Quito. Additionally, the consequences of spatial segregation are identified based on a literature review and empirical evidence regarding socioeconomic variables such as employment, access to services, crime, and security of tenure. Finally, the specific public policies proposed are aligned with the main socioeconomic outcomes affected by spatial segregation.

PROBLEM DESCRIPTION Patterns of spatial segregation of the low-income population Since the 1970s, urban development in Quito has followed patterns of “urban sprawl”. Quito has expanded its territory from

THE PUBLIC SPHERE | 2017 ISSUE

5,189 hectares to 43,550 hectares between 1971 and 2011 (Instituto de la Ciudad, 2013). As Figure 1 shows, the expansion of urbanization since 1970 has followed “leapfrog patterns of development.” Districts like San Antonio and Pomasqui in the north of the city, Calderon in the northeast, and Cumbaya and Tumbaco in the east – that were located in areas far from the urban perimeter – started to grow in 1987. As a result, the city has begun to resemble a haphazard patchwork, consuming more land than contiguous developments (Gillham & MacLean, 2002:4). Remaining open tracts were filled with new developments over time. Urban sprawl in Quito has been accompanied by the violent valorisation of urban land in central areas (Carrion et al., 1987: 84). The higher cost of land in city centres can be attributed to two factors: clustering and access (Gillham & MacLean, 2002:10). “Clustering,” also known as economies of agglomeration, implies that businesses benefit economically from trade if they are located in the same geographical space. Additionally, employees benefit from the ability to search for jobs in the same cluster, and employers benefit from the resulting labour pool (Gillham & MacLean, 2002:10). “Access,” on the other hand, means that people who live and work in city centres have better access to markets, basic services, and facilities than residents of peripheral areas. Spatial differences in terms of the commercial value of urban land in Quito are evidenced in Figure 2. The price of a square metre of land in peripheral districts such as San Antonio, Pomasqui and Calderon is between USD 5 and USD 47, whereas the price in central districts such as Iñaquito and Mariscal Sucre can be as high as USD 930 per square metre (Instituto de la Ciudad, 2016). As a response to the high costs of land in the city centre, low-income people have been forced to live in peripheral areas. These areas are characterised by an abundance of land, low provision of services, and limited access to employment, which justifies the low cost of housing. Figure 3 illustrates how low-income

[Addressing spatial segregation in Quito, Ecuador]

FIGURE 1 MAP OF URBAN GROWTH IN QUITO 0 2.25 4.5

13.5

18 Kilometres

Map key Year

1: 200,000

1983

District boundary

1760

1987

Rural area

1888

1995

1921

2003

1946

2006

1956

2011

1971

Source: Instituto de la Ciudad, 2016. Elaboration: Author.

FIGURE 2 MAP OF PRICE OF URBAN LAND IN QUITO (USD PER SQUARE METER)

1: 200,000 0 2.25 4.5

13.5

18 Kilometres

Map key Price (USD/sqm) 5-24 25-47 48-71 72-101 102-141

142-191 192-270 271-419 420-658 659-930

District boundary Rural area

Source: Instituto de la Ciudad, 2016. Elaboration: Author.

THE PUBLIC SPHERE | 2017 ISSUE

FIGURE 3 MAP OF POVERTY RATE IN QUITO 2010 (UNSATISFIED BASIC NEEDS) N

1: 200,000 0 2.25 4.5

13.5

18 Kilometres

Map key Poverty rate 12%-19%

30%-42%

20%-29%

65%-88%

43%-64%

Rural area District boundary

Source: National Institute of Statistics and Censuses (INEC), 2010. Instituto de la Ciudad, 2015. Elaboration: Author.

households in Quito have been segregated to peripheral areas while high and middle-income classes live concentrated around the central business district. According to the Unsatisfied Basic Needs method, the percentage of poverty in Quito is 30 per cent (INEC, 2010a). However, the analysis of the spatial distribution of poverty shows important differences between Quito’s districts. Centrally located districts, such as those located between Carcelen in the north and Mariscal Sucre in the south, have poverty rates from 12 to 19 per cent. The poverty rate of surrounding districts such as Pomasqui and Calderón increases up to 29 per cent. The most remote areas, such as Checa and Yaruqui, show poverty rates of up to 88 per cent. Hence, the more remote the district is from the city centre, the higher its poverty rate is.

[Addressing spatial segregation in Quito, Ecuador]

Consequences of spatial segregation Spatial segregation affects economic, social, and psychological dimensions of low-income inhabitants’ lives. The distance from employment centralities to peripheral neighbourhoods limits their job opportunities. The time spent in daily commuting reduces the amount of time they could spend on other activities, which constrains their capacity to overcome poverty. The quality of education in segregated neighbourhoods is likely to be lower than in central areas, which deepens inequalities in the long run. The establishment of informal settlements in the periphery also presents negative effects since residents have to dedicate time and effort to protect their property, which reduces their labour supply and income. The concentration of poverty in peripheral neighbourhoods also generates social problems such as crime, violence, and addiction. Lastly, segregation has psychological impacts on the poor, as it reduces their self-confidence and limits their aspirations.

Economic consequences The concentration of employment in the city centre limits employment opportunities for the segregated low-income population. Figure 4 shows how employment is concentrated in one specific central district of Quito: Iñaquito. According to the National Economic Census (INEC, 2010b), the businesses located in Iñaquito employ 110,341 people or 7,356 people per square kilometre. On the other hand, businesses located in peripheral districts such as La Ecuatoriana, Guamaní, and Turubamba (in the south) employ between 150 and 290 people per square kilometre. The distance from peripheral areas to employment centres represents a barrier to the low-income population, who must commute long distances to get to work and to access new employment opportunities.

THE PUBLIC SPHERE | 2017 ISSUE

Commuting long hours every day increases the phenomenon of time poverty. The time spent in commuting prevents citizens from dedicating more time to work, study, rest, or family activities. In other words, this reduces the poor’s “full income capability”, which is “the income an individual could generate if he or she spent all available time in paid work” (Burchardt, 2008:12). In Quito, an average trip between the peripheral areas and the central business district can take between one and two hours by bus. In effect, the impoverished are constrained to generate income to overcome poverty as they waste two to four hours of paid work by commuting. Segregation also affects the capacity of the poor to join employment networks and learn about job opportunities. Many residents in isolated areas of the city seldom travel to other neighbourhoods not only because they lack resources but also because they lack reasons, skills, and confidence (Kintrea, 2008:73). The disadvantaged “consider that the city centre is out of their bounds” (Kintrea, 2008:73). Consequently, they lack employment networks beyond peripheral areas and are less likely to learn about opportunities located in the city centre. Low-quality education is another reality for segregated low-income residents. Most high-quality educational facilities are concentrated in central districts. Therefore, if disadvantaged students want to access high-quality education, they have to commute long distances. Additionally, when poor families are clustered geographically and there are few schools close to poor districts, disadvantaged students will also be clustered geographically. This phenomenon generates lower performance of impoverished students in comparison with those who study in central areas because schools in poor areas have greater needs regarding the quality of teachers, infrastructure and educational resources. Given the importance of human capital for future opportunities, lower-quality education for the poor perpetuates

[Addressing spatial segregation in Quito, Ecuador]

FIGURE 4 MAP OF NUMBER OF EMPLOYEES BY DISTRICT IN QUITO N

1: 200,000

0 2.25 4.5

13.5

18 Kilometres

Map key Number of employees

0-602 603-2,838 2,839-8,008

8,009-14,452 14,453-31,374 31,375-58,653 58,654-110,341

District boundary Rural area

Source: National Institute of Statistics and Consensus (INEC), 2010. Elaboration: Author.

poverty and is likely to contribute to widening income inequality over time (Levy 1995, as cited in Squires, 2002:65). Spatial segregation leads the low-income population to establish informal settlements in peripheral areas. These communities are “settlements of the urban poor developed through the unauthorised occupation of land” (Huchzermeyer & Karam, 2006:vii). In Quito, the low-income population establish informal settlements as a mechanism to access low-cost urban land (Mena, 2010:6). The low prices offered by clandestine land developers to impoverished people are mainly related to the legal insecurity of land, the lack of basic infrastructure and services, and its location in peripheral and disaster prone areas (Clichevsky, 2003:10). Living in informal settlements demands time and effort from impoverished people, which reduces the household’s supply of labour. The establishment of these unauthorised communities is time-consuming because inhabitants have to self-manage

THE PUBLIC SPHERE | 2017 ISSUE

land, housing, services, and community equipment (Fernández, 2004:23). Additionally, people that live in informal settlements spend time and resources maintaining tenure security through informal means such as informal policing, community organizations and staying close to home. This reduces the household’s supply of labour and thus income (Field, 2007:1562). To sum up, being established in informal settlements reduces the possibilities for low-income people to overcome poverty.

Social and psychological consequences The concentration of the poor in peripheral areas leads to the concentration of social problems. As Rusk (1999) states, “poor neighbourhoods are poverty machines” (Squires, 2002:67). Residents of poor areas are frequently exposed to crime and counterproductive activities, which generates a cycle that perpetuates crime. The “street culture” of the poorest neighbourhoods is characterised by violence, drugs, sex, teenage pregnancy, and other problematic behaviours (Squires, 2002:65). According to Crane (1991), when people, especially teenagers, are exposed to negative behaviours in poor areas, they have a high probability of emulating these behaviours, even after controlling for the influence of a person’s immediate family and personal characteristics (Squires, 2002:64). Isolation of the poor from different contexts and social classes leads them to underestimate their abilities and limit their aspirations. Neighbourhoods are the areas in which beliefs, attitudes, and expectations are constructed through social relationships (Kintrea, 2008:72). Psychological and anthropological research has shown that the conditions of poor neighbourhoods generate “mental models” in low-income people, which limits their capacity to imagine a better life (The World Bank, 2015:14). These “neighbourhood effects” can be mitigated if impoverished households share spaces with mid-

[Addressing spatial segregation in Quito, Ecuador]

dle and high-income populations, with different aspirations and expectations. Social interactions of poor people in mixed communities can increase important development outcomes such as school achievement and labour market participation.

POLICY OPTIONS The essence of spatial segregation is the interaction of housing systems and labour market derived inequalities. These are therefore the most relevant areas of intervention (Kintrea, 2008:69). The proposed policies should guarantee the “right to the city” for the poor, which involves not only the right to access the centrality and its movement but also the right to inhabit, work, and freely use time and places in the city (Lefebvre, 1970, as cited in Mitchell, 2012:18). Finally, the proposed policies must be comprehensive in the sense that they should address the economic, social, and psychological effects of spatial segregation. Policy options will be assessed under the framework of “growth management” as the optimal approach to address uneven development in cities. Growth management refers to the attempts to use planning, policy and regulatory techniques to influence the allocation of new developments in cities (Gillham & MacLean, 2002:155). The main goal of growth management is to achieve “smart growth” in cities. The Smart Growth Network (2015) proposes the construction of “compact cities” as a way to achieve smart growth. This implies the manipulation of urban size and structure in pursuit of the environmental, social, and economic benefits derived from the concentration of urban functions. Compact cities are characterised by high population density, large built areas and intensive urban, economic and social activities (Burgess, 2000:9). Additionally, smart growth requires promoting

THE PUBLIC SPHERE â&#x20AC;&#x2020;| â&#x20AC;&#x2020;2017 ISSUE

the most effective use of existing land and infrastructure, establishing restrictions on development in outlying suburban areas and building more affordable housing in metropolitan areas (Squires, 2002; Smart Growth Network, 2015).

Provision of infrastructure and services in peripheral areas Urban sprawl and spatial segregation are important concerns for local policy makers not only because of their consequences for disadvantaged populations, but also because these phenomena present challenges in terms of provision of infrastructure and services. As more residential housing is built in peripheral areas, local governments need to increase their spending to build more schools, hospitals, water and sewage lines and electrical utilities (Williams, 2000:15). This is a suboptimal pattern of urbanization since the city continues expanding its urban territory instead of using efficiently the urban areas that are already equipped. Quito is a clear example of an uncontrolled urban expansion that has led to the inefficient use of land. According to the Metropolitan Plan for the Land Management of Quito, there are 86,448 properties without construction in the urban area, equivalent to 21 per cent of this area. Additionally, 83 per cent of these unused properties have access to water, sewer, and electricity services, 15 per cent have access to at least two of these services, and only 2 per cent of these properties does not have access to any basic service. Finally, 6 per cent of vacant land in Quito is public property, which is equivalent to 401 hectares. (STHV-MDMQ, 2011:17). Figure 5 shows the distribution of vacant land across Quito. Urban sprawl in Quito is characterised by low population density and suboptimal use of space. The current population density in Quito is 4,400 people per square kilometre (Demographia, 2016). This figure is relatively low compared with other Latin American capital cities such as Bogota (16,900 people/

[Addressing spatial segregation in Quito, Ecuador]

FIGURE 5 MAP OF VACANT LAND IN QUITO

Map key Vacant land Urban land

Source and Elaboration: STHV-MDMQ, 2011.

km2), Lima (11,900 people/ km2), and Mexico City (9,800 people/ km2) (Demographia, 2016). Furthermore, the allocation of individual families in single-family homes on individual lots indicates suboptimal use of space (Gillham & MacLean, 2002:7). In Quito, 99 per cent of the 634,611 existing homes are single family homes (INEC, 2010c). There are approximately 392,000 lots in Quito, with an average of 1.6 homes per lot. Therefore, infrastructure and services are not being used efficiently since the current population could be accommodated in a smaller area.

THE PUBLIC SPHERE â&#x20AC;&#x2020;| â&#x20AC;&#x2020;2017 ISSUE

The provision of necessary services and infrastructure to the entire population should be a priority for local policy makers. However, the optimal strategy should not be to continue investing in the provision of services for new developments generated as a result of uncontrolled urban expansion. Instead, the strategy should be the creation of a compact city, taking advantage of the available vacant land, infrastructure, and services to include the segregated population in the labour and housing dynamics of central areas of the city.

Regularization and relocation of informal settlements The establishment of informal settlements and the consequences derived from this form of occupation are important issues when analysing spatial segregation of the poor. There are two solutions in addressing informal settlements in cities: regularization and relocation. The regularization programs are aimed at formally integrating these settlements in the urban area and providing services to the newly developed land (Clichevsky, 2003:13). The relocation programs are normally applied in contexts in which it is not possible to establish long-term housing for the poor in these territories; for example, when informal settlements are located in disaster prone areas. Regularization of informal settlements represents clear advantages for their residents. The most important advantage is that occupants gain land tenure security through the incorporation of the property into the formal market. Additionally, regularised neighbourhoods are provided with infrastructure, basic services, and facilities. Both security of tenure and service provision are basic conditions to improve the welfare of the segregated population. Informal neighbourhoods that participate in regularization projects should meet certain conditions, which differ between

[Addressing spatial segregation in Quito, Ecuador]

projects and countries. Common requirements include that residents should have lived a minimum amount of time in the lot and that they do not own another property in the country (Clichevsky, 2003:17). It is important to point out that â&#x20AC;&#x153;regularizationâ&#x20AC;? does not mean that the piece of land is given for free to the occupants, but that the government frequently sells it at low or subsidised prices that can be paid in different instalments. Therefore, a common requirement in regularization programs, which often undermines the elimination of poverty, is to possess an income that allows some payment by the occupants (Clichevsky, 2003:17). Quitoâ&#x20AC;&#x2122;s urban sprawl has been accompanied by the establishment of informal settlements since the 1970s. The most notable regularization processes started in 2001. In the period between 2000 and 2010, 90 per cent of the new developments were built without the approval of the local authority. Half of these new developments were built in informal settlements (Cueva, 2011:2). In 2001, the local government counted 480 informal neighbourhoods and started the regularization process (Cueva, 2011:2). The process of regularization implies the delivery of property rights for the land and the implementation of required infrastructure, public spaces, risk mitigation projects, basic services, and facilities. By 2015, the local government had regulated 315 informal neighbourhoods (El Telegrafo, 2013, MDMQ, 2014:30, 2015:27). Although regularization of informal settlements increases the quality of life of low-income residents, regularization programs have some disadvantages. These initiatives can perpetuate poverty and segregation since the regularised lots will remain located in peripheral zones, away from employment centres and high-quality public facilities. The new legal neighbourhoods will also be exposed to the consequences of the concentration of poverty discussed in the previous sections, with a high probability of becoming neighbourhoods prone to social problems. Moreover, local governments face significant costs derived from the provi-

THE PUBLIC SPHERE â&#x20AC;&#x2020;| â&#x20AC;&#x2020;2017 ISSUE

sion of basic services, transport, and public facilities for the new legal developments. Finally, despite the fact that formalizing existing housing is cheaper for the government than building new houses for the poor, the conditions of the dwellings built in informal settlements are often unacceptable as human settlements for a decent life (Smolka, 2003:5). These low-housing conditions remain when the neighbourhoods are legalised. The other option is to relocate informal settlements, which implies the provision of housing to people in central areas of the city. In the case of Quito, some illegal neighbourhoods have not received the approval to be regularised because they were established in risks zones or ecological protection zones. In these cases, the authorities decided which households would be reallocated based on their socioeconomic characteristics (Cueva, 2011:1). Given the considerable extension of vacant land in Quito, the low population density, and the suboptimal use of space, the relocation of informal settlements towards central areas represents the optimal solution in favour of the creation of a compact and inclusive city.

Provision of infrastructure and services in peripheral areas The provision of social housing in central areas of the city has the potential to address the main consequences of spatial segregation of the low-income population. Social housing comprises the establishment of publically funded dwellings typically let at sub-market rents to disadvantaged populations (Robinson, 2012:2). The principal objective of these projects is to ensure decent housing for people that are most in need or struggling with their living costs (Flouri et al., 2015:2). The main advantage of social housing is that it provides stability, security and affordability in terms of accommodation for disadvantaged people. Additionally, the quality of publicly provided dwellings is usually

[Addressing spatial segregation in Quito, Ecuador]

significantly higher than the quality of houses that impoverished tenants could afford in the private sector (Hills, 2007:18). Although social housing projects around the world have achieved the objective of providing affordable and secure houses for the poor, they have presented some disadvantages. These projects generally constitute the poorest neighbourhoods of the city, prone to social problems that reinforce poverty conditions. As shown by Flouri et al. (2015:2), neighbourhoods with a high concentration of social housing have high levels of crime, unemployment, antisocial behaviour, low levels of educational attainment, and mental health problems. Therefore, some projects may be isolating the poor and their problems. Some social housing initiatives have failed to consider the strong impact that neighbourhoods have on the personal development of an individual in their design process. The â&#x20AC;&#x153;social contagion hypothesisâ&#x20AC;? states that neighbourhood influences, especially in relation to problems such as antisocial behaviour among older children, are transmitted through peers (Ingoldsby et al., 2006 as cited in Flouri et al., 2015: 1). Therefore, people exposed to antisocial behaviour in their neighbourhood are more likely to engage in delinquency and other counterproductive activities. Additionally, research has shown that neighbourhood contexts that make parents more exposed to stressful events may adversely affect their and their childrenâ&#x20AC;&#x2122;s mental health (Linares et al., 2001 as cited in Flouri et al., 2015:2). To avoid problems derived from the concentration of poverty in neighbourhoods, social housing projects should be built in mixed communities where low-income people interact with those from different backgrounds and income levels. It is important to consider that once a social housing project is established inside a middle or high-income neighbourhood, there is a possibility that higher-income tenants decide to move to avoid adverse effects related to crime and counterproductive behaviours. Therefore, the design of social housing projects should include strat-

THE PUBLIC SPHERE | 2017 ISSUE

egies to retain higher-income tenants in their neighbourhoods. These strategies could comprise improving neighbourhood conditions and services and increasing crime control in these areas (Hills, 2007:8; Kintrea, 2008:75). Additionally, investment in high-quality public schools in close proximity to social housing projects is key for attracting people from different income levels (Joseph and Feldman, 2009:623). Furthermore, the impact of neighbourhood effects is greater during childhood and adolescence (Galster et al., 2007:724); therefore, social housing projects should include public spaces in which children and teenagers from different backgrounds can interact. This will also contribute to creating bonds between these groups, disincentivising higher income families from leaving the neighbourhood. As a result of the presence of residents from different income levels, low-income tenants are expected to improve their educational outcomes, employment, and earnings (Department for Communities and Local Government, 2010:23). Some social housing projects have failed to serve as an effective welfare service because they have generated dependency rather than social mobility (Robinson, 2012:1). Disadvantaged people might see housing provided by the government as a guaranteed and time-unlimited service, which can reduce incentives to increase their income to purchase a house in the future. Therefore, these projects have the potential to undermine self-sufficiency and reduce incentives to look for a paid work (Robinson, 2012:2). Finally, tenants do not experience the consequences of their behavioural and financial actions because they cannot lose their homes if they become indebted and default. Consequently, social housing encourages the poor to continue with a “poor payment record and anti-social behaviour to the detriment of the wider community” (Robinson, 2012:4). Taking into account these factors, the design of new social housing projects must include mechanisms to promote social mobility and compliance with obligations by the target population.

[Addressing spatial segregation in Quito, Ecuador]

Although the quality of social housing is typically higher in comparison with the quality of houses that tenants could afford in the private sector, certain projects have failed to meet quality standards for “decent houses”. For instance, rates of overcrowding are higher in social housing projects than in other tenures in the United Kingdom; additionally, social tenants are likely to report dissatisfaction with their housing quality (Hills, 2007:3). Therefore, it is important that social housing initiatives guarantee “decent housing,” which means that dwellings must be designed according to household size and composition (Robinson, 1979:56-57). These houses must also provide “a reasonable state of repair, a degree of thermal comfort, modern facilities and services, and minimum levels of fitness for habitation” (DCLG, 2006 as cited in Morrison, 2013:2570). Quito’s local government has implemented nine social housing projects throughout the city since 2009. These projects are aimed at selling 7,100 houses to low and middle-income populations at a subsidised price. Moreover, the projects seek to reallocate families that were living in informal settlements established in risk zones. Social dwellings are located both in peripheral areas (2,600 houses in the north and 3,000 houses in the south) and in the historical centre (1,500 houses) (Municipio de Quito, 2013, 2014). According to the residents of social housing projects in Quito, these initiatives have represented an important change in their way of living. However, there are obstacles that prevent the poorest households from benefitting from these interventions. Zoila Perez, a resident of one project in the north, states: “I cannot describe in words the excitement that I feel when I arrive at my house, knowing that it is going to be here forever, and I will not be evicted.” There are two basic requirements to have access to social housing projects in Quito: a) To have savings equal to 10 per cent of the total value of the property; b) To be beneficiaries of the housing bonus provided by the government or to have

THE PUBLIC SPHERE | 2017 ISSUE

access to credit in any banking institution. It is unlikely that the poorest satisfy these requirements as they do not have savings nor access to credit. Therefore, these projects are mainly inhabited by middle-income families. Finally, most social housing projects in Quito are located in peripheral areas and have not made an effort to integrate social classes, keeping the poor isolated in poor neighbourhoods. Social housing projects located in the centre of the city have the potential to address the main consequences of spatial segregation of the low-income population. However, the success of these projects is conditional on certain features of their design. First, these projects should consider the characteristics of the neighbourhood that surrounds them. These neighbourhoods should not concentrate poverty and social problems, but instead should be composed of households from different backgrounds and income levels. Additionally, the design of these projects should include complementary strategies to retain higher-income tenants in communities. Social housing projects should also include mechanisms to avoid generating dependency and promote social mobility, such as the establishment of obligations and conditions for tenancy. The dwellings provided must meet quality standards to be considered “decent housing.” Finally, these projects should be affordable for the poorest inhabitants, which implies the implementation of complementary programs that provide access to credit or certain concessions for rent payments.

Inclusion of segregated population into the labour market Strategies to overcome the concentration of unemployment in peripheral neighbourhoods include the promotion of diversity and the provision of services to increase jobs opportunities for the disadvantaged. The creation of mixed communities, in which people from different social backgrounds and professions

[Addressing spatial segregation in Quito, Ecuador]

interact, will result in workless tenants interacting with employed people. This, in turn, will increase predisposition to work and job options for the unemployed (Robinson, 2008:106). Advice regarding opportunities in the labour market and assistance in job seeking, especially for young people, has increased employment among social tenants (Hills, 2007: 18). Social housing programs could also include the construction of training centres to strengthen skills and capacities of job-seekers. It is also important to promote initiatives aimed at facilitating tenantsâ&#x20AC;&#x2122; work attendance, such as the implementation of childcare support. Moreover, employment opportunities created by the public sector should prioritise hiring low-income residents. Another option is the attachment of conditions to social tenancies to encourage new tenants to improve their employability (Stephens, 2008:36). Commitment contracts in social housing projects are usually focused on responsibilities and obligations such as specific standards of conduct to tackle antisocial behaviour (Robinson, 2008:107). Similarly, these contracts can include the obligation to prove that unemployed tenants are continually seeking employment. Finally, once tenants are successfully allocated in the labour market, there is a high probability that they desire to become property owners (Kintrea, 2008:78). If they are renting a dwelling provided by the government, they may look for the possibility of buying a house. If the conditions of the social housing project do not allow residents to buy the house, it is likely that they would leave the project to purchase their own property in another area. Therefore, it is important that social housing initiatives provide some flexibility in the type of ownership that allows tenants to upgrade their status and stay in the project once they have a higher level of income. This will increase positive neighbourhood effects in the community. In summation, social housing projects must be comprehensive in the sense that housing policies must be complemented by

THE PUBLIC SPHERE | 2017 ISSUE

inclusive employment programs. These interventions comprise the provision of advice, training, and assistance in job seeking for social tenants. Additionally, the creation of mixed communities, the provision of child care support, and the prioritisation of social tenants for jobs created by the government will increase tenants’ opportunities. Social housing projects should also create mechanisms for tenants to upgrade their type of ownership - from renters to owners, for example - as their income increases.

CONCLUSION AND POLICY RECOMMENDATIONS Conclusion The analysis of the economic, social, and psychological consequences of the spatial segregation of the poor has provided the basis to identify the primary goals for required interventions. Policy makers should guarantee access to employment opportunities and high-quality education in central areas of the city for disadvantaged people. The poor should also have access to housing in the city centre to avoid commuting long hours, which will increase the time they can spend on productive activities such as work or studies. It is necessary to upgrade informal settlements to reduce the time and resources that they spend stocking and protecting their homes. Additionally, it is important to reduce the concentration of the poor in specific neighbourhoods to avoid the proliferation of social problems and counterproductive activities. Finally, the interaction of the low-income population with people from different social backgrounds is essential to extend their employment networks, mitigate social problems, and increase their educational and employment aspirations. Optimal policies to address spatial segregation of the poor should be designed under a smart growth framework, contribut-

[Addressing spatial segregation in Quito, Ecuador]

ing to the construction of compact cities maximizing efficiency in resource use. The provision of infrastructure and services to the entire population must be a priority for policymakers. However, the optimal strategy to address spatial segregation is not to continue investing in service provision in the periphery as the city grows uncontrollably, but to take advantage of existing vacant land, infrastructure and services to build a compact city in the current urban area. The same reasoning applies to the treatment of informal settlements located in peripheral areas, for which the best option is not their regularization but their relocation towards central areas. The analysis shows that the optimal strategy to address spatial segregation of the poor is the construction of comprehensive social housing projects in central areas of the city.

Policy recommendations Spatial segregation of the low-income population can be addressed through the implementation of comprehensive social housing projects, which should be in line with the following guidelines: ďż˝ Target low-income populations: These projects should serve low-income inhabitants living in peripheral areas of the city, including residents of legal properties and informal settlements. ďż˝ Central location: Social housing should be located in central areas to guarantee access to employment, high-quality education, services, and facilities for the poor. These projects should take advantage of vacant land and existing services and infrastructure. ďż˝ Mixed-income neighbourhoods: Social housing should be accompanied by the construction of neighbourhoods where people from different backgrounds and levels

THE PUBLIC SPHERE | 2017 ISSUE

of income interact. Additionally, it is necessary to implement strategies to retain higher-income tenants in social housing neighbourhoods, which comprises improving neighbourhood conditions and services, increasing crime control in the zone, investing in high-quality public schools inside the neighbourhood, and building public spaces in which children and teenagers can interact. These interventions are expected to reduce the concentration of social problems and improve educational outcomes, employment, and earnings. � Promote social mobility: Social tenancy contracts should include conditions related to time limits, specific standards of conduct, and the obligations for the unemployed to prove that they are continually seeking employment. These conditions will promote selfsufficiency and decrease the dependency of social tenants. � Affordability: The dwellings provided must be accessible for the poorest inhabitants. Therefore, they should not require initial payments or access to private credit. Instead, they should be progressive; starting with sub-market rents that increase as a tenant’s income rises. The public sector can also increase the affordability of social housing by providing credit under special conditions such as lower interest rates, flexible instalments, and the absence of collateral requirements. � Quality standards: Dwellings must meet minimum standards to be considered “decent housing,” which includes: design according to household size and composition, a reasonable state of repair, a degree of

[Addressing spatial segregation in Quito, Ecuador]

thermal comfort, modern facilities and services, and minimum levels of quality for habitation. � Complementary employment programs: Social housing programs must be complemented with initiatives aimed at integrating the low-income population into the labour market. These interventions include the provision of ad vice regarding opportunities, assistance in job seeking, training to strengthen skills and capacities of job-seekers, implementation of childcare support, and hiring the low-income population for jobs created by the government. � Flexibility in tenure: Social tenants should have the opportunity to decide whether to rent or buy properties. Tenants that start renting the dwellings should be given enough flexibility to buy their properties once their income increases.

REFERENCES Burchardt, T. (2008). Time and income poverty. London: Centre for Analysis of Social Exclusion -London School of Economics. Retrieved 10 October 2015, from http://eprints. lse.ac.uk/28341/1/CASEreport57.pdf Burgess, R. (2000). The Compact City Debate: A Global Perspective. In Jenks, M., & Burgess, Rod. Compact cities: Sustainable urban forms for developing countries. London; New York: E. & F.N. Spon Carrión, D. et al. (1987). El Proceso Urbano en Ecuador. Quito: Instituto Latinoamericano de Investigaciones Sociales (ILDIS). Carrión, F. (1993). La cuestión urbana en Quito. Retrieved 15 November, 2015 from http:// works.bepress.com/cgi/viewcontent.cgi?article=1085&context=fernando_carrion Clichevsky, N. (2003). Pobreza y Acceso al Suelo Urbano: Las políticas de regularización en América Latina. Guatemala: CEPAL-AECI. Crane, J. (1991). Effects of Neighborhood on Dropping Out of School and Teenage Childbearing. In the Urban Underclass, edited by Christopher Jenks and Paul E. Peterson (299-320). Washington D.C.: Brookings Institution. Cueva, S. (2011). Las Deficiencias de la Regularización. Cuatro Barrios de Quito. Quito: Instituto de la Ciudad.

THE PUBLIC SPHERE | 2017 ISSUE

DCLG (Department of Community and Local Government) (2006). A decent home: definition and guidance for implementation. Retrieved 02 April, 2016, from http://www. communities.gov.uk/publications/housing/decent home Demographia (2016). Demographia World Urban Areas. 12th Annual Edition: 2016:04. Retrieved from http://www.demographia.com/db-worldua.pdf Department for Communities and Local Government (2010). Evaluation of the Mixed Communities Initiative Demonstration Projects. London: Crown. El Telégrafo. (2013). La Unidad “Regula Tu Barrio” está en las Ocho Administraciones Zonales: 227 Barrios Ilegales ya Están Regularizados. Retrieved 20 November, 2015 from http://www.editogran.com.ec/noticias/quito/item/moradores-de-227-barrios-ya-cuentan-con-escrituras.html Fernández, R. (2004). Pobreza y proceso de urbanización. Curso Latinoamericano de Especialización en hábitat y pobreza. Módulo 2. Centro de Estudios Políticos Económicos y Sociales CEPES - Centro de Estudios en Hábitat y Pobreza Urbana. Buenos Aires. Field, E. (2007). Entitled to Work: Urban Property Rights and Labor Supply in Peru. The Quarterly Journal of Economics, 122(4), 1561-1602. Flouri, E., Midouhas, E., & Tzatzaki, K. (2015). Neighbourhood and own social housing and early problem behaviour trajectories. Social Psychiatry and Psychiatric Epidemiology, 50(2), 203-213. Galster, G., Marcotte, D., Mandell, M., Wolman, H., & Augustine, N. (2007). The Influence of Neighborhood Poverty During Childhood on Fertility, Education, and Earnings Outcomes. Housing Studies, 22(5), 723-751. Gillham, O.; MacLean, A. (2002). The limitless city: a primer on the urban sprawl debate. Washington, DC: Island Press. Greenstein, R.; Sabatini, F., and Smolka, M. (2000). Urban Spatial Segregation: Forces, Consequences, and Policy Responses. Land Lines: November 2000, Volume 12, Number 6. Harvey, D. (2003). The right to the city. International Journal of Urban and Regional Research, 27(4), 939-941. Hills, J. (2007). Ends and Means: The Future Roles of Social Housing in England. Retrieved 10 November, 2015, from http://eprints.lse.ac.uk/5568/2/Ends_and_Means_The_future_roles_of_social_housing_in_England_2.pdf Huchzermeyer, M. & Karam, A. (2006). Informal Settlements: A Perpetual Challenge? Cape Town: UCT Press. INEC - Instituto Nacional de Estadísticas y Censos. (2010a). Pobreza por Necesidades Básicas Insatisfechas – NBI. Retrieved 14 November, 2015 from http://www.ecuadorencifras.gob.ec/pobreza/ INEC - Instituto Nacional de Estadísticas y Censos. (2010b). Base de Datos del Censo Nacional Económico. Quito: INEC. INEC - Instituto Nacional de Estadísticas y Censos. (2010c). Base de Datos del Censo de Población y Vivienda. Quito: INEC. Ingoldsby, E., Shaw, M., Winslow, D., Schonberg, S., Gilliom, E., & Criss, M. (2006). Neighborhood Disadvantage, Parent–Child Conflict, Neighborhood Peer Relationships, and Early Antisocial Behavior Problem Trajectories. Journal of Abnormal Child Psychology, 34(3), 293-309. Instituto de la Ciudad. (2013). Boletín Conociendo Quito No. 3. Quito: Instituto de la Ciudad. Instituto de la Ciudad. (2016). Mapa del Precio del Suelo urbano en Quito. Quito: Instituto de la Ciudad.

[Addressing spatial segregation in Quito, Ecuador]

Joseph, M., & Feldman, J. (2009). Creating and Sustaining Successful Mixed-Income Communities. Education and Urban Society, 41(6), 623-652. Kintrea, K. (2008). Social Housing and Spatial Segregation. In Fitzpatrick, S., Stephens, Mark, & Shelter. The future of social housing. London: Shelter. Lefebvre, H. (1970) (Translation in 2003). The urban revolution. Minneapolis: University of Minnesota Press. Levy, F. (1995). The Future Path and Consequences of the U.S. Earnings/Education Gap. Economic Policy Review 1(1): 35-41. Linares, L., Heeren, T., Bronfman, E., Zuckerman, B., Augustyn, M., & Tronick, E. (2001). A Mediational Model for the Impact of Exposure to Community Violence on Early Child Behavior Problems. Child Development, 72(2), 639-652. Mena, A. (2010). Regularización de los asentamientos informales en Quito: Análisis de las políticas públicas. Unpublished Msc. Thesis, Latin American Faculty of Social Sciences, Quito. Retrieved 27 November 2015, from http://repositorio.flacsoandes.edu. ec/bitstream/10469/2383/8/TFLACSO-2010APMS.pdf Mitchell, D. (2012). The Right to the City: Social Justice and the Fight for Public Space. New York: Guilford Press. Morrison, N. (2013). Meeting the Decent Homes Standard: London Housing Associations’ Asset Management Strategies. Urban Studies, 50(12), 2569-2587. Municipio de Quito. (2013). Conozca los planes de vivienda social que impulsa el Municipio de Quito. [Video]. Retrieved 20 November 2015 from https://www.youtube.com/ watch?v=x_3dkXeImTk Municipio del Distrito Metropolitano de Quito (MDMQ) (2014). Informe de Rendición de Cuentas del Alcalde y Municipio de Quito 2014. Quito: MDMQ. Municipio del Distrito Metropolitano de Quito (MDMQ) (2015). Informe de Rendición de Cuentas del Alcalde y Municipio de Quito 2015. Quito: MDMQ. Oxley, M., & Institute for Public Policy Research. Forum on the Future of Social Housing. (2000). The future of social housing: Learning from Europe. London: Institute for Public Policy Research. Robinson, D. (2008). Social housing and worklessness. In Fitzpatrick, S., Stephens, Mark, & Shelter. The future of social housing. London: Shelter. Robinson, D. (2012). Social Housing in England: Testing the Logics of Reform. Urban Studies. 50(8) 1489–1504, June 2013. Robinson, R. (1979). Housing Economics and Public Policy. London: Macmillan. Rusk, D. (1999). Inside Game, Outside Game: Winning Strategies for Saving Urban America. New York. Brookings Institution Press. Smart Growth Network. (2015). What is smart growth? Retrieved 21 November 2015 from http://www.smartgrowth.org/what-is-smart-growth/ Smolka, M. (2003). Informalidad, pobreza urbana y precios de la tierra. Land Lines, Enero 2003, Volumen 15, número 1. Retrieved 20 November, 2015 from http://nute.ufsc.br/ moodle/biblioteca_virtual/admin/files/bibliot._partic._aula_2_[3].pdf Squires, G. (2002). Urban sprawl: Causes, consequences, & policy responses. Washington, D.C.: Urban Institute Press. Stein, D., & Valters, C.. (2012). Understanding theory of change in international development. Justice and Security Research Programme, International Development Department, London School of Economics and Political Science. Stephens, M. (2008). The role of the social rented sector. In Fitzpatrick, S., Stephens, Mark, & Shelter. (2008). The future of social housing. London: Shelter.

THE PUBLIC SPHERE | 2017 ISSUE

STHV-MDMQ. Secretaría de Territorio, Hábitat y Vivienda del Municipio del Distrito Metropolitano de Quito. (2011). Plan Metropolitano de Ordenamiento Territorial 20122022. Quito: Municipio de Quito. The Centre for Theory of Change (2013). What is Theory of Change? Retrieved 10 October 2015, from http://www.theoryofchange.org/what-is-theory-of-change/ The World Bank (2015). World Development Report: Mind, Society, and Behaviour. Washington, DC: The World Bank. United Nations. 1996. Report of the United Nations Conference on Human Settlements (Habitat II) Istanbul: United Nations. Whitehead & Scanlon. 2007. Social Housing in Europe. London School of Economics and Political Science. Retrieved 20 November 2015 from http://www.lse.ac.uk/geographyAndEnvironment/research/London/pdf/SocialHousingInEurope.pdf Williams, D. (2000). Urban sprawl: a reference handbook. Santa Barbara, Calif.: ABCCLIO.

Why have UK pension funds not taken on social impact investing? Dana Elman Vishkin

Master of Public Administration in Public Policy and Management, Class of 2016 London School of Economics and Political Science

ÂŠThomas Hawk

Photo modified to black and white

ABSTRACT According to the UK government and Big Society Capital, pension funds are not investing enough in social and environmental impact investments. Pension funds are the largest asset owners in the UK economy and are legally obligated by fiduciary duty, to maximise financial returns. Many impact investment practitioners argue that fiduciary duty is the main reason holding back pensions from making impact investments, for various reasons. In this research based on my masterâ&#x20AC;&#x2122;s dissertation, I argue that fiduciary duty is not the most critical reason holding back pension funds from impact investing. I designed a case study on UK pensionsâ&#x20AC;&#x2122; engagement with the impact investment market to test my hypothesis that measuring the impact of an investment leads to high transaction costs and is a limiting factor.

[Why have UK pension funds not taken on social impact investing?]

INTRODUCTION The UK’s social impact investment1 market is considered a success among G8 countries. However, the UK government and market champion Big Society Capital (BSC) argue that the market has yet to achieve scale with institutional investors: “The UK is still a distance from having a robust and scalable social investment market but has played a leading role in much of the innovation in social impact investment internationally” (UK National Advisory Board [NAB] Report to G8 Taskforce, 2014, p.27). Many perceive that institutional investors, especially Pension Funds (PFs)2, have a role in financing the global transition to a socially positive and climate resilient economy (Della Croce et al, 2011; ClearlySo, 2009). This is often justified because the volume of PFs’ Assets Under Management (AUM) has the power to make a significant impact on the global economy. PFs are the largest asset owners in the economy and are legally obligated to fiduciary duty, to maximise financial returns to their beneficiaries.3 Many argue that fiduciary duty is the main barrier for PFs to invest in the market. Other obstacles have been acknowledged in literature. The UN’s Principles of Responsible Investment (PRI) organisation and others have been advocating for a new paradigm: not only that Environmental, Social and Governance (ESG) risks do not jeopardise fiduciary duty, but also that they should be weighted in investment decisions because positive social impact and good financial performance are linked (PRI, 2012). In 2015 the Law Commission advocated that trustees (fiduciaries) should explain whether ESG risks are considered prior to making investments. However, the UK government declined to accept the Law Commission’s recommendation, leaving many practitioners disappointed (Williams, 2015).

1. Social Impact Investments are “those that intentionally target specific societal and/ or environmental objectives along with a financial return and measure the achievement of both” (UK National Advisory Board to G8 Taskforce, 2014, p.5).

2. PF is a common asset pool aimed “to generate stable growth over the long term, and provide pensions for workers when they reach retirement” (ClearlySo, 2011).

3. A legal duty commonly known as imposing obligations on trustees or fiduciaries to maximise investment returns to their beneficiaries.

THE PUBLIC SPHERE | 2017 ISSUE

4. In 2015, some of the biggest investment managers globally signalled that their clients find interest in the market, among them are BlackRock (US-headquartered, with second largest office in the UK) and Goldman Sachs (Mudaliar et al., 2016)

Despite the fiduciary duty barrier, some UK PFs are engaging with the impact investment market through collaborative initiatives of impact investments. At the same time, global investment managers have begun signalling that their clients (i.e. PFs) are interested in the impact investment market.4 Observing both BSC’s argument that PFs have not taken on impact investing, and institutional investors’ positive efforts to engage with the market, I am interested in whether PFs in the UK have taken part in impact investing, and if not, why not? This study discusses my findings on UK PFs’ engagement with the market.

LITERATURE REVIEW

5. The National Association of Pension Funds (NAPF) former name is PSLA (the Pensions and Lifetime Savings Association).

Research shows that in 2009, UK PFs had $2.5tn (around £1.7tn) of AUM, which was estimated as 9 per cent of the world’s total AUM (ClearlySo, 2011). This sum is larger than UK’s GDP and is spread between two types of PFs, public pensions and private pensions, which are different in nature. Public PFs are grouped by a system of 89 funds. Together they comprise the £179bn UK Local Government Pension Scheme (LGPS) in England and Wales (Flood, 2016). The Local Authority Pension Fund Forum (LAPFF) brings together 70 LGPS funds from across the country with combined assets of over £175bn (LAPFF website). Private funds, but not exclusively, are grouped in two leading lobbying organisations: the Association of Member Nominated Trustees, and the National Association of Pension Funds.5 The latter represents 1,300 PF member funds that provide pensions for more than 17m people. NAPF owns more than £900bn of assets (NAPF website). Recent UK policy changes will have a significant effect on increasing personal contributions for pensions. By 2018, automatic enrolment of workers to a retirement

[Why have UK pension funds not taken on social impact investing?]

fund should lead to nine million new pension savers (Galdiolo et al, 2015). Therefore, with the AUM growth expected from automatic enrolment, PFs’ assets are estimated to grow by more than £600bn by 2030 (Keohane and Rowell, 2015).

Initiatives by pension funds to engage with social impact investment market Over the recent years, PFs have begun to co-invest through collaborative initiatives. Some of these collaborations have been considered responsible and sustainable investments, and most recent initiatives relate to impact investing, for example, the Capital Market Climate Initiative (CMCI)6 and the Investing for Growth (I4G) initiative. The latter is the major initiative analysed in this study. The I4G consortium was launched in March 2013 by five local government PFs, including Greater Manchester Pension Fund (GMPF), West Yorkshire Pension Fund, West Midlands Pension Fund, South Yorkshire Pension Fund and Mersyside. This initiative was built on the growing trend of divesting from unethical investments, such as alcohol producers, and aimed to generate positive impact (Anderson, 2015). I4G brought together PFs which jointly own around £30bn in investible assets (Lokhandwala, 2014). It reflected an innovative investment trend in channelling retirement funds to generate positive social or environmental impact. This initiative will be discussed in length in chapter five.

6. CMCI brings experts to help deliver private climate financing at scale in developing countries (OECD, 2011).

Pension funds’ barriers to impact investments Many barriers to UK PFs seeking to engage in the impact investment market were found in a large study prior to the launch of I4G initiative (Smith Institute, 2012. P.6). These challenges include: 1. A trade-off between social impact and financial return, which is perceived to contradict the fiduciary duty to

THE PUBLIC SPHERE | 2017 ISSUE

maximise returns for investors. 2. Potential local conflicts of interest may occur if the investments are linked to local or geographical schemes. 3. PFs invest with strategy for a long-term investment horizon strategy. Hence, the impact investments targets that mainly interest PFs are infrastructure projects for their attractive long-term income streams. 4. There are reputation risks associated with new investments. Overall, fiduciary duty is commonly considered the most important reason that PFs avoid impact investing and keep to mainstream financial investments.

Fiduciary duty

7. Risk-Adjusted Return is “a concept that refines an investment’s return by measuring how much risk is involved in producing that return, which is generally expressed as a number or rating. When comparing two or more potential investments, an investor should always compare the same risk measures to each different investment in order to get a relative performance perspective.” (Investopedia)

The UK’s institutional environment is considered strong, where investors, creditors and clients are legally protected to exercise their rights (Rojas-Suarez, 2014). Fiduciary duty is a legal and economic mechanism for ensuring that clients’ best interests are kept by their investment managers. This is a fundamental institutional tool to ensure trust across UK’s capital market actors. Similarly, the disclosure principle is an important financial accounting mechanism. It enables standardisation of financial information reporting, and its goal is communicating financial activities clearly and accurately. When institutional investors consider engaging with a growing capital market, such as the impact investment market, they are expected to be prudent and cautious, behaving to ensure their fiduciary duty is maintained. PFs are generally considered more conservative than other institutional investors because of their hesitation to innovate with their beneficiaries’ assets. Due to PFs’ fiduciary duty and responsibilities, they focus on investments with at least a market risk-adjusted financial return (Drexler and Noble, 2013). 7 PFs’ hesitation from engaging with impact investing arises from a perceived trade-off between financial return and social impact (Grabenwarter & Liechtenstein, 2011). Social impact is

[Why have UK pension funds not taken on social impact investing?]

a concept that brings uncertainties to the investment field for several reasons; it takes time to achieve social change and measure the investment’s social impact. Proving that there are impact returns, i.e. that the social change happened and it was the cause of the specific investment, requires an analysis of the investment’s counterfactual. Finding the answer to the most critical question “what would have happened in the absence of the investment” requires monitoring data on various characteristics of the investee. More importantly, an answer to the question proves whether there is a causal link between the investment and the environmental or social change that the investee delivered. These actions result in a costly investment strategy where PFs face unfamiliar challenges. Impact measurement alone is challenging to conduct. Hence, I assume it is difficult for PFs to engage with this investment strategy if they do not have the knowledge or expertise to conduct impact measurement.

There is a positive link between fiduciary duty and delivering social or environmental change The UN’s PRI organisation and others have been advocating for a new paradigm: not only that ESG risks do not jeopardise fiduciary duty, but that they should be weighted in investment decisions because positive social impact and good financial performance are linked (PRI, 2012). PFs seek comfort and low risk (Friedman, A. Public lecture at the LSE Alternative Investment Conference. 18th January 2016), and impact investing allows PFs to diversify their investment portfolio with specific assets often uncorrelated with other assets. Another justification for a logical link between fiduciary duty and social change is that delivering social impact requires time (normally several years), in line with PFs’ long-term horizon. Practically, one of the few PFs that invest in UK’s growing impact investment market is Greater Manchester PF. It argues that

THE PUBLIC SPHERE | 2017 ISSUE

its fiduciary duty does not prevent it from investing in northern communities in the UK, and that impact investing locally encourages economic participation and creates social and environmental impact (GMPF annual report, 2014). My study focuses on GMPF because it leads other UK PFs in the I4G initiative, setting an example with its commitment to impact investing.

Other common barriers More barriers that hold back investors from impact investing are acknowledged by the Global Impact Investing Network (GIIN) annual survey on impact investing, Nesta in the UK, the Global Social Entrepreneurship Network and the G8 taskforce: (1) the ecosystem is currently considered fragmented, which suggests its development is hindered (Nesta website, 2015), (2) there is a lack of sufﬁciently mature impact ventures with proven business models and revenue streams for investors to work with (GIIN, Nesta & the Global Social Entrepreneurship Network), and (3) a lack of a favourable legal environment (G8, 2014).

Potential solutions from overseas to overcome barriers Recent OECD research on the global impact investment market (OECD, 2015) finds that when some of the US’ largest asset owners invested in impact investments, they were often subsidised by the government (Wood et al, 2012). US case studies reveal that when the government or philanthropists create grants to provide investors with catalytic capital offering “first loss” funding, they guarantee an agreed protection ratio in case of a “first loss” to the capital committed. This “first loss” policy tool has been widely used by the US administration to incentivise growth by small business and VC growth (EY, 2014); it has been supporting US impact investors as well (interview with Social Finance UK, October 28, 2015). However, although “first

[Why have UK pension funds not taken on social impact investing?]

loss” capital is a mechanism that reduces the risk of the investment, it is not sustainable for a high volume of investments. This is not the case in Australia, where, in 2015, an Australian PF named HESTA committed AUD $30m to impact investments targeting competitive returns (Mudaliar et al., 2016). HESTA reveals on its website that it awarded the assets to Social Ventures Australia (SVA), a not-for-profit organisation focused on overcoming the disadvantages in Australia (HESTA, 2016). SVA manages the Social Impact Investment Trust, one of Australia’s largest impact investment allocations. It represents the largest single commitment by an Australian superannuation fund to the local impact investing market. HESTA recently closed its first investment, a loan of AUD $6.5m to Horizon Housing, a non-profit community housing provider which operates properties and projects across 15 local government areas in Queensland and northern New South Wales (HESTA, 2016). France has also set an example to other countries with its Solidarity Investment Fund initiative. The Solidarity Investment Fund directs 10 per cent of each saver’s assets to social investments, and 90 per cent to traditional companies that qualify as socially responsible (Keohane and Rowell, 2015). According to The Financial Times, French Solidarity Investment Funds have been investing this way for more than a decade with more than €4.8bn AUM (Cumbo, 2015). The UK NAB report recommended following the French model, and since 2014, specific UK market leaders have been advocating for the endorsement and adoption of social investment pension options (UK NAB, 2014). One prominent supporter is former UK NAB Chairman, also BSC’s co-founder and former CEO, Nick O’Donohoe. He now chairs the UK’s New Dormant Account Committee and argues the UK should obligate employers to offer an option for “Defined Contribution pension” savers where they pay a portion of their retirement fund into a social fund (Cumbo, 2015).

THE PUBLIC SPHERE | 2017 ISSUE

UK government regulations to encourage impact investments

8. Financially material investment – “Not all ESG information is created equally in the eyes of the investor. An investor considers any factor that can have a significant impact on an investee’s (company’s) core business value drivers – namely growth, profitability, capital efficiency and risk exposure – to be financially material.” (RobecoSAM definition).

The UK government has been encouraging impact investments through several changes in regulation since 2002 (UK NAB report to G8, 2014). In 2014, with the efforts of BSC, the Social Investment Tax Relief (SITR) was introduced to provide a 30 per cent tax relief for social investments (Palin, 2015). Additionally, in 2014, the Investment Intermediaries Fiduciary Duties Reform was launched by the Law Commission, aiming to address legal uncertainty and lack of clarity around fiduciary duties (UK NAB, 2014). In 2015, two departments asked the Law Commission to specifically review the UK Equity Markets and Long-Term Decision Making for identifying how fiduciary duties apply to investors. The Law Commission recommended that PFs, usually focused solely on maximising financial returns to their beneficiaries, should take into account ethical or ESG factors, with one condition: to ensure that investments are financially material.8 Despite the government’s refusal of the Law Commission’s recommendations (Williams, 2015), I understand that fiduciary duty is not compromised when institutional investors engage with impact investments.

Literature review conclusions The UK regulatory environment is more supportive today than ever before for institutional investors to consider social impact investing. However, there is wide uncertainty around fiduciary duty and impact investments. Despite hesitations, there are initiatives for PFs, led by PFs, that engage with impact investing in joint-investments. In contrast to the French PFs, UK PFs face a regulatory disadvantage, because UK beneficiaries do not have a demand mechanism for requesting their PF managers to engage with this type of investment.

[Why have UK pension funds not taken on social impact investing?]

ARGUMENT AND THEORY I argue that fiduciary duty is not the most critical reason holding back pension funds, and test the hypothesis that measurement fees are a limiting factor because they lead to high transaction costs. I hypothesise that PFs invest less when transaction costs of impact measurement are high. I test my hypothesis by designing a case study on UK PF engagement with the market: H0: There is no relationship between impact measurement costs to PFsâ&#x20AC;&#x2122; engagement with social impact investing. The commonly perceived relationship is between fiduciary duty and engaging with this investment strategy. As the latter is perceived to generate below-market financial returns, PFs donâ&#x20AC;&#x2122;t want to jeopardise their fiduciary duty, and hold back from it. H1: PFs invest less when transaction costs of impact measurement are high. My argument is that transaction costs of impact measurement are high due to impact investmentsâ&#x20AC;&#x2122; impact measurement requirements, and the outsourcing of the due diligence from PFs to asset managers (i.e. intermediaries). PFs often hire services of intermediaries to manage their investments, and consequently, pay them for management fees. These management fees are the reward that PFs pay to their fiduciaries. For an impact investment to take place, transaction costs should be at prices that reflect supply and demand. Furthermore, engaging in impact investments may come with an additional price tag because an impact investment deal requires impact-measurement knowledge and experience. To gain these, I expect that PFs would use the same intermediaries who man-

THE PUBLIC SPHERE | 2017 ISSUE

age their portfolios for deploying capital into the new impact investing strategy. Now I turn to dissect impact investment and measurement costs: A PF may find that developing this unique knowledge is challenging and costly because impact investment is a new playing field. This investment strategy offers more uncertainty and complexity than mainstream “pure-financial-return” investments. Finding the appropriate social enterprise to invest in takes time, resources and exceptional professional skills. This set of skills is not in the PFs’ main line of business. By applying the transaction cost theory (Coase theorem) (Calabresi, 1968) to explain why PFs hold back from impact investing, I hypothesise that there is a market failure, leading UK PFs to limit or avoid impact investing because trading costs are very high:

9. The specific nature of the impact investment intermediary is not significant for this study, for example, whether PFs pay to a mainstream financial intermediary (e.g. KKR), or a specialised impact investment fund (e.g. Bridges Ventures). Once invested in an impact investment deal, the deal’s costs are considered internalised.

�

Impact investments bear information costs for PFs in researching potential social enterprise investees. These are higher than researching traditional investments because they are outside PFs’ expertise.

�

Contracting costs are higher in impact investments than for mainstream investments due to the requirement of establishing specific terms to deliver social change in addition to financial performance. For PFs, contracts to allocate and commit assets are written only after a profound evaluation and risk analysis which are mandatory due to their fiduciary duty and responsibilities.

These impact measurement costs lead to higher fees.9

[Why have UK pension funds not taken on social impact investing?]

Implications of impact investments: Outsourcing the dealâ&#x20AC;&#x2122;s impact measurement, as part of the diligence process, to a specialised intermediary does not ensure low transaction costs because closing a deal takes resources. As explained by the principal-agent theory, when a PF (principal) requires an intermediary (an agent) to reach a goal, a misalignment of incentives and objectives emerges. The intermediary exercises authority on behalf of the PF and incurs the costs of its efforts. Through its activities, the intermediary may be better informed than the PF about the true costs of the impact investment. This theory explains that asymmetric information exists where parties with different objectives have different sources and access to information. In an impact investment deal, the use of an intermediary may lead to adverse selection,10 due to information asymmetry benefiting the intermediary. As explained by transaction cost theory, when PFs decide to invest in an impact investment deal, fees paid to the intermediary should reflect the additional impact-measurement capability provided. In this research I expect that UKâ&#x20AC;&#x2122;s largest PFs (in Part A of my research), and GMPF (part B), would invest through an intermediary. In the I4G initiative (part B), I expect to find that sharing the due diligence costs, including impact measurement costs, would result in an outcome where joint-investing is preferable over independent investing, because collaboration between investors leads to reduced costs and risks, in contrast to individual investments (I4G website, 2015).

10. A concept in economics which captures the following idea: intermediaries with better private information about the quality of a product will selectively participate in trades which benefit them the most, at the expense of their pension funds clients (source: Wikipedia).

RESEARCH DESIGN These are the two building blocks for my research design:

THE PUBLIC SPHERE | 2017 ISSUE

Part A: Assessment of the gap between ideal and actual volume of impact investments I conduct interviews with I4G stakeholders and practitioners. All interviews were conducted from February 2016 to May 2016. I first assess the ideal impact investment volume that the UK government and practitioners have sought to achieve, by asking each interviewee (Table 1) the following questions: � �

“What do you think would constitute “appropriate scale” for the UK impact investment market, and why?” “Do you think the UK impact investment market reached appropriate scale (as a sustainable financial market)?”

Secondly, I identify the barriers that, in policymakers’ and investors’ view, hold back pension funds. I asked each interviewee: “what obstacles remain to further scale PFs’ investments to impact portfolios?” Table 1 List of interviews

11. Anonymity of this investor would improve my research ability to clearly reflect the investment manager’s words.

Organisation name

Name of interviewee and position

Investing for Growth (I4G) Initiative

Brian Bailey, I4G Initiative Facilitator

UK Government, Cabinet Office

Robert Hewitt, Senior Policy Advisor, Social Investment & Finance Team

Bill & Melinda Gates Foundation

Nick O’Donohoe, Senior Advisor, Chairman of the New Dormant Account Committee and former BSC CEO

Bridges Ventures, UK

Emilie Goodall, Director of Impact+

A London-based £13.5bn investment management firm

Managing Partner, co-Head of Private Equity of a leading financial intermediary investing on behalf of PFs11

Source: Author.

[Why have UK pension funds not taken on social impact investing?]

Thirdly, I research the market’s actual volume by capturing investments made by UK’s 20 largest pension funds in 2015 (Deane-Williams, 2015).12 I examine the first 20 through identifying the volume of assets they committed in 2015. For identifying an investment as an impact investment, I use Bridges Ventures’ Impact+ conceptual framework that differentiates between investments’ types (Bridges Ventures, 2015) (Figure 1).

12. Full list is in Appendix A online at publicspherejournal. com.

FIGURE 1 SPECTRUM OF CAPITAL

Spectrum of Capital Financial Only

Responsible Sustainable

Impact only

Impact

Delivering competitive financial returns Mitigating Environmental, Social and Governance (ESG) risks Pursuing Environmental, Social & Governance Opportunities

Focusing on measurable high impact solutions

Focus

Limited or no regard for environmental, social or governance (ESG) practices

Mitigate risky ESG practices in order to protect value

• PE firm

Examples

integrating ESG risks into investment analysis

• Ethically

screened investment fund

Adopt progressive ESG practices that may enhance value

• Best-in-Class SRI fund

• Long-only

public equity fund using deep integration of ESG to create additional value

Address societal challenges that generate competitive financial returns

Address societal challenges where returns are as yet unproven

• Publicly listed • Social fund dedicated Impact to renewable energy projects Bonds/ Development (e.g. a wind farm) Impact • Microfinance Bonds structured debt fund (e.g. loans to microfinance banks)

Address societal challenges that require a below-market financial return

Address societal challenges that cannot generate a financial return

• Fund providing quasi equity or unsecured debt to social enterprises or charities

Source: Bridges Ventures, 2015.

Part B: Greater Manchester pension fund case study’s design I design a case study to analyse GMPF’s impact investments during its leadership in I4G initiative, the joint investment vehicle of five local government pension funds. From 2013 to 2015, GMPF invested jointly through its leadership in I4G. After learn-

THE PUBLIC SPHERE | 2017 ISSUE

ing I4G did not continue to a further joint-investment, I search for barriers that PFs faced in I4G through primary data analysis.

THE GAP BETWEEN ACTUAL AND IDEAL VOLUME OF IMPACT INVESTMENTS Ideal volume of institutional impact investments My analysis firstly demonstrates that there is no monetary target for the UK impact investment market, resulting in a policy void. The government and market players provide different views on what the ideal volume of impact investments is, hence it is currently subjective. I find that with no unity and clarity for what constitutes ideal volume to achieve scale of institutional impact investments, it is challenging to claim that the market has yet to achieve scale. This is due to the following findings:

13. There is no official UK definition for the term ‘social economy’. However, according to UK Social Economy Alliance, the term includes organisations, social enterprises, co-operatives, universities, housing associations, crowdfunders, social investors, think tanks and charities.

A. The UK government does not have a monetary target for the volume of institutional impact investments. In addition, the government’s designated unit for social investment does not track the market’s scalability. When the Cabinet Office was in charge of this unit, until summer 2016, it fostered the demand for capital among social enterprises, through developing and pushing forward the ‘social economy’ vision.13 B. There was no attempt by the government to monitor the market’s volume of institutional impact investments. C. Measures for assessing scalability are subjective. When I asked ‘what do you think would constitute “appropriate scale” for the UK impact investment market’, interviewees provided diverse answers from different parallel industries. Bridges Ventures, The

[Why have UK pension funds not taken on social impact investing?]

Cabinet Office and Nick O’Donohoe agree that it is challenging to define “appropriate volume” for institutional investment in UK market.

Views on barriers to institutional impact investments I identify the barriers that, in policymakers’ and investors’ view, hold back pension funds:

Barrier 1: Impact investments suffer from weak branding with PF managers. Many PFs think that impact investing, as an investment strategy, contradicts fiduciary duty. However, there is wide evidence from Bridges Ventures and Generation fund’s financial returns that show this perception is false. Barrier 2: Lack of incentives to engage with the industry is a major barrier. Barrier 3: “Impact investing is very difficult to do” – says the co-head of a £13bn PE, and a Principles of Responsible Investment (PRI) signatory. The firm provides financial services to UK’s largest PFs, and the latter do not request the firm to engage with impact investing at all. Moreover, the interviewee clearly states that he has no incentive to change the firm’s investment practices. Being a ‘responsible investor’, i.e. examining ESG concerns as obligated by PRI membership, is sufficient. Barrier 4:: Investment managers, who invest on behalf of PFs, seek investment performance with a shorter horizon than the typical

THE PUBLIC SPHERE | 2017 ISSUE

social impact investment’s long-term horizon. Delivering social change requires a long-term investment horizon of a few years, which may contradict managers’ short-term performance. Most often, investment managers think about their performance according to periodical reporting, while impact investing requires longer timescales. This gap is explained by the principal-agent theory.

Pension funds’ actual volume of impact investments in 2015 I research the market’s actual volume and capture impact investments made by UK’s 20 largest pension funds in 2015. I find that only four out of these 20 PFs allocated assets to impact investments, as detailed in Figures 2 and 3. I find that the four PFs invested more than £1,300m into social or environmental impact investments in 2015. Together the four PFs that invested in 2015, committed together £1,380m in assets as shown in Figure 3. I also find that 13 of the 20 PFs report on outsourcing their asset management responsibility to institutional investment advisers, i.e. intermediaries, who are giant investment firms. Three intermediaries’ names often appeared: BlackRock Investment Advisors (UK) Limited, L&G and Towers Watson. BlackRock and L&G have recently been engaging in impact investing, and further research on their impact investments might shed light on PFs’ investments. Further details on UK BlackRock’s and L&G’s asset allocation to impact investments in Appendix B (online).

Data limitations and challenges in identifying which investment is truly an impact investment: When I analysed the 20 PFs’ annual reports, I found that reporting and disclosure principles for impact investments are not

[Why have UK pension funds not taken on social impact investing?]

FIGURE 2 TOP UK PENSION FUNDS (TOTAL AUM, GBP MILLIONS, 2015) AND SHARE OF IMPACT INVESTMENTS OF AUM Universities Superannuation Scheme

1.43%

BT Pension Scheme BTPS

1.16%

£49,500 £43,084

Royal Bank of Scotland Group Pension Barclays Bank UK Retirement Fund (UKRF)

£33,529 £26,800

N/A £21,876

HSBC Bank (UK) Pension Scheme

£21,625

BP Pension Fund Greater Manchester Pension Fund (GMPF)

0.85%

£17,590

Strathclyde Pension Fund

£15,758 £15,445

Aviva Staff Pension Trustee (ASPT) Ltd

£13,990

British Steel Pension Scheme

£12,932

BAE Systems Pension Fund BBC Pension Fund

0.16%

£12,862

British Coal Pension Schemes

£9,571

Unilever Pension Fund

£7,677

British Airways Pension Fund

£7,509

Lloyds Bank Pension Scheme No. 2

£7,082

AkzoNobel CPS Pension Scheme

£2,921

Source: Analysis of PFs’ annual reports for the year 2015 only. Publicly available annual reports missing for large PFs including Electricity Supply Pension, National Grid PF, Barclays UKRF and Railways Pension Trustee Company. Lloyds Bank shows data only for PS No.2. FIGURE 3 PENSIONS’ ASSET ALLOCATION TO IMPACT INVESTMENTS (GBP MILLION), 2015

£710 £500

Universities Superannuation Scheme

BT Pension Scheme BTPS

£150

£21.2

Greater Manchester Pension Fund GMPF (Invest4Growth)

BBC

Source: Analysis from UK pensions’ financial reports.

THE PUBLIC SPHERE | 2017 ISSUE

14. See Appendix C (online at publicspherejournal. com) for detailed explanation of challenges in researching nonfinancial terminology in PFs’ annual reports

15. Attempts to conduct an interview with GMPF’s relevant director for Property and Local Investments, Paddy Dowdall, resulted unfruitful. 16. GMPF refers to impact investing as ‘value investing’ (GMPF’s AR, 2013)

standardised and lead to confusion in annual reports’ non-financial terminology.14 For example, when I analysed British Telecom Pension Scheme (BTPS) impact investment strategies, I needed to use more than just BTPS’s published annual reports.

GREATER MANCHESTER PENSION FUND CASE STUDY & ‘INVESTING FOR GROWTH’ I reviewed GMPF’s annual reports from 2013 to 2015 and analysed GMPF’s actual volume of impact investments through a quantitative breakdown.15 Data analysis of GMPF’s annual reports of the period from 2013 to 2015 reveals that the trustees acknowledge the growing UK impact investment market. GMPF consistently disclosed its long-term approach and management strategies, which found the case of impact investing16 compelling (GMPF, 2013).

Greater Manchester’s fiduciary duty GMPF’s long-term approach considers that “participation in economic growth is a major source of long term equity return… [GMPF acknowledges that impact investing (D.E.V)] may result in prolonged periods of over and underperformance in comparison to a style neutral approach” (GMPF, 2013, p.107). However, in 2015 GMPF reiterated that its engagement with impact investing seeks both commercial returns and positive local impact, without risking its fiduciary duty responsibility. GMPF began with impact investing by committing assets to I4G and I analyse its involvement through the initiative’s press releases and online coverage. Additionally, I present my findings from an interview with the I4G facilitator, Brian Bailey. I identify GMPF’s leadership during 2013-2014, and discuss the barriers that held back the consortium’s PFs from further engaging with

[Why have UK pension funds not taken on social impact investing?]

impact investing. My analysis also includes discussion of GMPF’s independent investment strategy in 2015, once it had left I4G.

Greater Manchester’s leadership in the ‘Investing for Growth’ initiative – 2013-2014 Table 2 presents the assets committed by I4G PFs to five impact funds, three of them managed by Bridges Ventures. The consortium invested £152m, mostly through Bridges Ventures’ funds, and diversified the investments to various sectors including property, SIBs, and the social sector. GMPF led the consortium to channel a total of £40m to two other funds: £25m was committed to Boost & Co. fund for industrial lending, and £15m to Midven Ltd. to an unknown goal. In both smaller funds, GMPF was the leading investor. Table 2 Asset commitment to impact investment by I4G PFs Pension fund

Bridges Venture Property Fund

Bridges Venture SIB Fund

Bridges Ventures Social Sector Fund III

Boost&Co Industrial Lending Fund 1

Midven LTD AM&M Fund

Total Committments

GMPF

£25m

£2.5m

£10m

£50m

MPF

£10m

£1m

£5m

WMPF

£30m

WYPF

£10m

£5m

SYPA

£10m

£5m

East Riding

£20m

Total

£105m

£17m £30m £5m

£20m £15m £20m

£3.5m

£25m

£15m

£152m

Source: I4G Concluding Statement, June 2014.

THE PUBLIC SPHERE | 2017 ISSUE

The absolute majority of I4G’s assets (69 per cent) was committed to property funds. This was in line with the I4G mandate to seek competitive financial returns: “Whilst meeting their investment return and risk strategies, …and to seek positive economic, social and environmental impact in the UK. The funds are not reducing their risk and return requirements for these investments” (I4G website).

Property investment is a typical strategy for institutional investors, as appears also from BSC’s main strategic operations. The rationale, according to BSC’s theory of change for social housing, is that sometimes social intervention models require a lot of capital if using property or capital-intensive assets. Social housing is affordable home funding for low-income people and those facing particular challenges in finding suitable housing (BSC, 2016). According to I4G facilitator, Brian Bailey (interview, May 1, 2016), GMPF was the most progressive PF among the five. The I4G initiative faced barriers in allocating PFs’ assets to impact investments. The programme was terminated and has not been followed by a further joint initiative. Impact investing is now only being used by a few funds in very specific areas, and GMPF continues to invest in impact investing.

Barriers to I4G initiative I identify several barriers that held back the consortium’s pension funds from following-up with further collaborative initiatives: (1) current opportunities offer small size deals and below the average investments; (2) trustees and fiduciaries’ workload in other areas has greater priority; (3) market and risks are not well understood; (4) investment consultants for pension funds do not offer impact investment services due to the amount of evaluation work involved, and to the perception that these investments can

[Why have UK pension funds not taken on social impact investing?]

only have negligible impact on overall investment returns; (5) no external pressure exists (from beneficiaries) on funds to look at impact investments. I find these barriers to limit the scalability of UK’s impact investment market.

Greater Manchester’s investment strategy analysis 2013-2015 When the consortium initiative was not renewed in 2015, GMPF independently made impact investments, without other pension funds, and committed approximately £150m in 2015. At 31 March 2015, GMPF’s total AUM was £17,590m. Of this total, £17,316m was held in the ‘Main Fund’ and invested across a broad spread of assets, while £275m was held in the ‘Designated Fund’ and invested wholly in UK index-linked gilts and cash (GBMP, 2015). GMPF published its Core Belief (CB) Statement on its website in 2009, setting out the key underlying beliefs in relation to investment issues and GMPF’s overall approach to investment matters. The beliefs mentioned in GMPF’s CB statement provide the rationale to explain its investment activity, that “allocations to asset classes other than equities and government bonds (e.g. corporate bonds, private equity and property) are positively considered as they offer the fund other forms of risk premia (e.g. additional solvency risk/illiquidity risk),” (CB Statement, 2009, p.3).

GMPF’s net assets and impact portfolio evolution In 2015, GMPF disclosed investment information through the fund’s annual report, where it reported on both strictly financial and impact-related investments. However, the fund’s impact and non-financial objectives were limited in details, while financial reporting (of net assets and asset allocation to the fund’s Local Investment portfolio) was clear, as shown in Figure 4.

THE PUBLIC SPHERE | 2017 ISSUE

FIGURE 4 GREATER MANCHESTER PENSION FUND NET ASSETS (TOTAL AUM, GBP MILLIONS), 2013-2015

£17,590 £12,589

£13,284

2013

2014

2015

Source: Annual reports, 2013-2015.

Asset allocation to impact investments Over the past three years, GMPF reported its allocation of assets to impact investing through complementing strategies: 1. Local Investment: Engagement in the joint initiative I4G as one part of its ‘local investment’ portfolio, focused within the Greater Manchester area. 2. Impact portfolio: In 2015, GMPF disclosed that due diligence has been undertaken on potential opportunities for impact investing and listed the following impact themes: renewable energy, SME loans and social infrastructure (GMPF, 2015).

1. Local investment portfolio In 2013-2014, GMPF supported local investment through its Property Venture Fund, where its target allocation ranged up to 3

[Why have UK pension funds not taken on social impact investing?]

per cent of the Main Fund, together with other allocations. GMPF’s Property Venture Fund invests within the North West of England, focusing specifically on the Greater Manchester area in (1) property development and (2) redevelopment opportunities. GMPF’s Property Venture Fund has increased its allocation to local investment (a type of impact investment) each year. Based on GMPF’s annual report analysis, Figure 5 below demonstrates the Property Venture Fund’s change of investments flows into impact investment (GMPF, 2013-2015). FIGURE 5 GMPF’S ALLOCATION TO LOCAL INVESTING THROUGH PROPERTY VENTURE FUND (GBP MLNS)

£86 £42

2013

£57

2014

2015

Source: Analysis of GMPF annual reports’ analysis, 2013-2015.

In March 2014, GMPF approved to allocate £50m, a share of 0.38 per cent of its total AUM to the I4G initiative via its Property Venture Fund. GMPF was the largest participant in I4G, and allocated the largest volume. Figure 6 on the next page outlines the timeline of GMPF’s asslocation to I4G, from its local investment portfolio.

THE PUBLIC SPHERE | 2017 ISSUE

FIGURE 6 DISCLOSURE OF GMPF’S ASSET ALLOCATION AND COMMITTMENT TO THE I4G INITIATIVE

March 2014: Allocated £50m to the initiative As of March 2015, committed £38m By 31 March 2015, £20m were drawn down and invested by the fund GMPF reported in March 2015: Managers are making satisfactory progress against the initial objectives

Source: Analysis of GMPF annual reports’ analysis, 2013-2015.

2. Impact portfolio I find that GMPF outsourced due diligence to Bridges Ventures’ Property Alternatives Fund III. This specific fund invests in direct property and “property-backed business across strategic sectors” (Bridges Ventures Website, 2016), and it achieved a final close in June 2015 with £212m of equity commitments, together with other former I4G partners - Merseyside, and South Yorkshire PFs.

CONCLUSIONS AND POLICY RECOMMENDATIONS Conclusions The argument that PFs are not investing enough in social and environmental impact investments, pushed forward by UK government and BSC, lacks evidence. £1,300m were already invested by PFs in 2015, and these were judged by the UK government as insufficient.

100

[Why have UK pension funds not taken on social impact investing?]

For testing this argument, more policy work is needed for setting desirable monetary milestones for the capital allocated and invested in the market. Hence, I conclude that the UK government has yet to complete its market development role. Only when the social investment market has a monetary target for achieving scale, would it then be justified to argue that it had not achieved that scale. Until then, it is upon subjective opinions to decide if the ÂŁ1,300m already invested by PFs led the market to achieve scale in 2015. My findings show that disclosure of non-financial terminology and asset commitments lacks clarity. Disclosure of non-financial terminology for impact investments should be standardised for achieving clearer reports on all investments, including those targeted for both financial and social impact. The lack of standards in reporting should be fixed. GMPF demonstrates that some pension funds have taken on social impact investing. Its Fiduciary duty is linked to participation in economic growth, and after the consortium ended, GMPF explored investment opportunities independently. My findings suggest that GMPF faced barriers in 2013-2014, which held it back from continuing investing with other PFs in the consortium. But in 2015, it overcame barriers and invested independently in impact investments. More importantly, when GMPF moved from a joint-investment to independent investment, it could not share the transaction costs with other PFs, but impact investments were continued even with the increased costs. When the consortium initiative was not renewed, GMPF independently invested ÂŁ150m in 2015. To further engage with impact investing, Greater Manchester had to solely incur costs of impact measurement when it individually invested. Based on these findings, GMPF internalised the required impact-measurement actions required for impact investment deals. It paid higher transaction costs when it invested independently in 2015, than it did before at the consortium. Additionally, these findings suggest

101

THE PUBLIC SPHERE | 2017 ISSUE

that for UK local government PFs, collaborating with other funds on impact investments may be an inferior strategy to independent impact investing. Furthermore, GMPF’s example demonstrates that not all PFs are yet to take on social impact investing. Some PFs link fiduciary duty to participation in economic growth, as a major source of long term equity return. Additionally, higher transaction costs did not stop GMPF from investing in 2015, after it left the consortium. Therefore, I find evidence that supports that H0 is not true, because GMPF is not holding back due to its fiduciary duty. Hence, fiduciary duty is not the most critical factor holding back PFs from impact investing. There must be other factors involved. However, I conclude that transaction costs are not a limiting factors when a PF is keen to engage with impact investing. Hence, I reject my hypothesis that high transaction costs limit PFs from impact investing. Through my findings, I discovered more potential factors that may explain why many pension funds still hold back from impact investing. I believe that more research will bring better understanding and open the discussion beyond relying on fiduciary duty.

Policy recommendations I recommend learning from GMPF’s positive experience, and addressing the barriers that held back PFs before in I4G. Some of these barriers are solvable in the short-term, while others are more challenging and require UK government and BSC to design a longer-term strategy. I recommend the following steps in Table 3 on the following page.

102

[Why have UK pension funds not taken on social impact investing?]

Table 3 Policy recommendations Short-term goals

Medium and long-term goals

Further educate market participants to understand risks.

If regulatory changes are feasible, create external pressure (from beneficiaries) on funds to look at impact investments: as with the French “Solidarity Funds”.

Incentivise intermediaries to offer opportunities that increase size deals to reach pensions’ average investments (i.e., learn from Greater Manchester’s preference to invest directly through an intermediary).

If regulatory changes are not feasible, incentivise asset managers who work on behalf of pensions to consider impact investing:

Address the perception among pension managers that impact investments provide weaker performance that traditional investments.

Incentivise investment consultants to offer impact investment services by:

1. Address the fact that asset managers often seek horizon returns in alignment with their investment performance reviews (shorter than impact investments’ horizons). 2. Provide incentives for trustees and fiduciaries to prioritise impact investments.

� Reducing the amount of evaluation work involved. � Improving perception that impact on overall returns is not negligible

REFERENCES Anderson, S. (2015, June 16). Divest and reinvest: 6 local governments that are innovating for sustainability. Retrieved 7 December, 2015, from http://platformlondon.org/2015/06/16/divest-and-reinvest-6-local-governments-that-are-innovating-for-sustainability/#sthash.a2owu1M0.dpuf Bell, M., and Paull, G. (2014). Is it about the money, money, money? Frontier Economics Bulletin. June 2014. Benjamin, B. (2015) Asset Management 2020: A brave new world. USA: PwC. Retrieved 7 February, 2016, from http://www.pwc.com/us/en/asset-management/investment-management/asset-management-2020.html Big Society Capital. (2014). Annual report. Theory of Change. Retrieved 1 May, 2016, from http://www.bigsocietycapital.com/sites/default/files/Theory%20of%20change%20 August%202015.pdf Big Society Capital. Social Housing. Retrieved 19 May, 2016, from http://www.bigsocietycapital.com/social-issues/housing-and-local-facilities

103

THE PUBLIC SPHERE | 2017 ISSUE

Big Society Capital. Scale. Retrieved 19 May, 2016, from http://www.bigsocietycapital.com/ what-we-do/our-strategy/scale Big Society Capital. Funding Affordable Homes. 27 January 2016. Retrieved 19 May, 2016, from http://www.bigsocietycapital.com/what-we-do/investor/investments/funding-affordable-homes Bridges Ventures. Property Alternatives Fund III Retrieved 12 May, 2016 from http://bridgesventures.com/property-fund/ Bridges Ventures. (2015) The Bridges Spectrum of Capital. Retrieved July 29, 2015 from http://bridgesventures.com/wp-content/uploads/2015/11/Spectrum-of-Capital-online-version.pdf Calabresi, G. (1968). “Transaction Costs, Resource Allocation and Liability Rules—A Comment”. Journal of Law and Economics 11 (1): 67–73 [p. 71–73] Cameron, D. (2010, July 19). Retrieved 17 March, 2016, from https://www.gov.uk/government/speeches/big-society-speech ClearlySo organisation (2011). Investor Perspectives on Social Enterprise Financing. Retrieved 17 April, 2016 from https://www.cityoflondon.gov.uk/business/economic-research-and-information/research-publications/Documents/research-2011/Investor%20Perspectives%20on%20Social%20Enterprise%20Financing.pdf Cross, L. (2016, April 7) Interview: BAE Systems pension fund talks housing Social Housing website Retrieved 12 May, 2016 from http://www.socialhousing.co.uk/interview-baesystems-pension-fund-talks-housing/7014714.article Cumbo, J. (2015, 2 September), “Use pension funds for UK social investment, ministers urged”. Financial Times. Retrieved 17 March, 2016, from https://next.ft.com/content/061b6b94-5165-11e5-b029-b9d50a74fd14 Dean-Williams, P. (2015). “P&I/ TW 300 Analysis, Year-end 2014”. UK: Towers Watson. Della Croce,R., C. Kaminker and F. Stewart (2011), “The Role of Pension Funds in Financing Green Growth Initiatives”, OECD Working Papers on Finance, Insurance and Private Pensions, No. 10, OECD Publishing, Paris. DOI: http://dx.doi.org/10.1787/5kg58j1lwdjd-en Drexler, M., and Noble, A. (2013). From the Margins to the Mainstream- Assessment of the Impact Investment Sector and Opportunities to Engage Mainstream Investors. Geneva: World Economic Forum. Flood, C. (2016, 10 January). Pooling UK local pension fund assets is ‘mere tinkering’. Financial Times. Retrieved 17 March, 2016, from https://next.ft.com/content/47fd8998-b5fb-11e5-b147e5e5bba42e51 Institutional Investors Lecture. (2016). [Public Speech by Friedman, A. Chief Executive Officer at GAM group.] Public lecture at the LSE Alternative Investment Conference. January 18, 2016 Freireich, J., and Fulton, K. (2009) Investing for Social and Environmental Impact. New York: The Monitor Institute. Retrieved 7 February, 2016, from http://monitorinstitute. com/downloads/what-we-think/impact-investing/Impact_Investing.pdf Galdiolo, S. and De Ste Croix, C. (2015). Responsible Investment Performance of UK Asset Managers The 2015 ShareAction Survey. Retrieved 27 April, 2016, from http://shareaction.org/wp-content/uploads/2016/01/AssetManagerSurvey2015.pdf Giddens, M. (January 18, 2016). Public Speech at the 2016 London School of Economics (LSE) Alternative Investment Conference. The Global Impact Investing Network. Retrieved May 18, 2016 from https://thegiin.org/ impact-investing/

104

[Why have UK pension funds not taken on social impact investing?]

Grabenwarter, U., and Liechtenstein, H. (2011) In search of Gamma – an unconventional perspective on impact investing. Madrid(?): IESE Business School. Retrieved 17 September, 2015, from http://socialvalueint.org/sources/iese-business-school/ HESTA. Your investments could be helping society and earning a return. Retrieved 15 May, 2016 from http://www2.hesta.com.au/update/member/impact-investing-feb2016/ Investing For Growth. Concluding Statement from June 2014. Retrieved 15 May, 2016 from http://www.lapfforum.org/press/files/Investing4growthI4GConcludingStatement12062014.pdf Investing for Growth Initiative. About us. Retrieved 7 December, 2015, from http://www. investing4growth.co.uk/Aboutus.html Investment & Pension Europe News. (September 11, 2015) National Grid in-house manager confirms sale to LGIM. Published by IPE Staff. Retrieved 12 May, 2016 from http:// www.ipe.com/news/asset-managers/national-grid-in-house-manager-confirmssale-to-lgim/10009766.fullarticle James, D., Kane, D., and Ravenscroft, C. (2016) Understanding the capacity and need to take on investment within the social sector. Retrieved 11 May, 2016, from https://www.ncvo.org.uk/images/documents/policy_and_research/funding/investment-within-the-social-sector-2016-summary.pdf Keohane, N. and Rowell, S. (2015). Good Pensions: Introducing social pension funds to the UK. Retrieved 27 April, 2016, from http://www.smf.co.uk/wp-content/uploads/2015/09/Social-Market-FoundationSMF-BSC-030915-Good-Pensions-Introducing-social-pension-funds-to-the-UK-FINAL.pdf Legal & General Group Website. Retrieved 12 May, 2016 from http://www.legalandgeneralgroup.com/ Marriage, M. (2016, April 24). Pension funds review tobacco divestment. Financial Times. Retrieved 17 May, 2016, from https://next.ft.com/content/e87a9b3c-0708-11e6-9b510fb5e65703ce Lokhandwala, T. (2014). UK local authority pension funds commit to social impact investments. Investment and Pensions Europe. Retrieved 12 May, 2016 from http://www. ipe.com/news/uk-local-authority-pension-funds-commit-to-social-impact-investments/10002168.fullarticle Mudaliar, A., Schiff, H., and Bass, R. (2016). 2016 Annual Impact Investor Survey Retrieved 22 May, 2016 from https://thegiin.org/knowledge/publication/annualsurvey2016 The National Association of Pension Funds (NAPF) website, Retrieved 22 May, 2016 from http://www.plsa.co.uk/ Nilsson, A. and Robinson, D. T. (2012). What is the business of business? (Research Report 90). Stockholm: SIFR the Institute for Financial Research. North, D. (1990). “A Transaction Cost Theory of Politics,” J. Theor. Politics, 2:4, pp.355-67. OECD Report: Social Impact Investment: Building the Evidence Base, 2015 Oxlade, A. (2015, September 12). Global list of biggest pensions exposes Britain’s flawed retirement plan. The Telegraph. Retrieved 27 April, 2016, from http://www.telegraph. co.uk/finance/personalfinance/pensions/11854493/Global-list-of-biggest-pensionsexposes-Britains-flawed-retirement-plan.html Palin, A. (2015, March 18). Budget 2015: Boost for social impact investing. The Financial Times. Retrieved 7 February, 2016, from http://www.ft.com/cms/s/0/88c9252a-cd7811e4-a15a-00144feab7de.html#ixzz3jHGim1k9 Pearce, B. (2014) Adapting and evolving Global venture capital insights and trends 2014. EY.

105

THE PUBLIC SPHERE | 2017 ISSUE

Robinson, M. (2015, January 28, 2016) Big Society Capital. Big Society Capital 2015 Numbers. Retrieved 12 May, 2016 from http://www.bigsocietycapital.com/latest/type/ blog/big-society-capital-2015-numbers Pension Funds’ Annual Reports: Annual reports of UK’s 20 largest PFs were drawn from online search. Rojas-Suarez, L. (2014). Towards Strong and Stable Capital Markets in Emerging Market Economies. (Papers No 75). Bank for International Settlements. Ronald Cohen Website. Retrieved 12 May, 2016 from http://www.ronaldcohen.org/about Saltuk, Y., Idrissi, A.E., Bouri, A., Mudaliar, A. and Schiff, H. (2015) Eyes on the Horizon: The Impact Investor Survey. New York: J.P. Morgan and Global Impact Investing Network. Retrieved September 7, 2015 from https://thegiin.org/knowledge/publication/eyeson-the-horizon Smith Institute (2012). Local Authority Pension Funds: investing for growth. The Smith Institute September 2012. Thinking About Impact, Part 1. (2016). [Video recording by Trelstad, B.]. MBA and Impact Investing Network Training (MIINT), NovoEd online platform. The Social Impact Investment Taskforce, established under the UK’s presidency of the G8. Impact Investment: The Invisible Heart of Markets. Retrieved 7 February, 2016, from http://socialimpactinvestment.org/reports/Impact%20Investment%20Report%20FINAL[3].pdf UK HM Treasury. (2012). A new approach to public private partnerships. Retrieved May 1, 2016 from https://www.gov.uk/government/uploads/system/uploads/attachment_ data/file/205112/pf2_infrastructure_new_approach_to_public_private_parnerships_051212.pdf UK National Advisory Board (NAB) report to the G8 Social Impact Investment Forum. (2014) Retrieved May 1, 2016 from https://www.gov.uk/government/groups/social-impact-investment-taskforce UKSIF mission and history. Retrieved 12 May, 2016 from http://uksif.org/about-uksif/ our-mission/ UN Principles for Responsible Investment Website https://www.unpri.org The UN Principles for Responsible Investment Initiative. Annual Report 2015 - From Awareness to Impact. Retrieved May 20, 2016 from https://www.unpri.org/download_report/9483 Wilson, R (2016). Social investment: a force for social change - UK strategy 2016. Retrieved 12 May, 2016 from https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/507215/6.1804_SIFT_Strategy_260216_FINAL_web.pdf Williams, J. (2015, November 13). UK decision not to amend rules on fiduciary duty ‘extremely disappointing’ Retrieved 27 April, 2016, from http://www.ipe.com/news/ esg/uk-decision-not-to-amend-rules-on-fiduciary-duty-extremely-disappointing/10010763.fullarticle Wood, D., Thornley, B., and Grace, K (2012): “Impact at Scale: Policy Innovation for Institutional Investment with Social and Environmental Benefit”, Pacific Community Ventures, Institute for Responsible Investment at Harvard University and Rockefeller Foundation, February.

106

Human trafficking and ethnic minority problems in Myanmar Policy recommendations for Myanmar and neighboring states Seunghyun Han

MA in International Relations and Economics, Class of 2016 Johns Hopkins University School of Advanced International Studies

ÂŠ Samantha Michaels

ABSTRACT Myanmar has gained a notorious reputation as a source and transit country for human trafficking within the Southeast Asian region. Human trafficking is not only prevalent internally, but trafficking across borders is sizeable as well. Ethnic minorities are particularly vulnerable to trafficking due to their conflict with the Burmese military government and the lack of minority rights within Myanmar. To analyse the connection between the human trafficking situation and minority conflicts in Myanmar, I will first examine the history of conflict between minority groups and the Myanmar government, the extent of human trafficking in Myanmar, and how human trafficking contributed to the militarisation of certain minority groups. This paper will then introduce key issues with the Myanmar government’s current policies, specifically those under the new National League of Democracy (NLD) administration, led by Aung San Suu Kyi. Finally, this paper makes the following policy recommendations: (i) develop effective measures to address the core grievances of ethnic minorities’ problems, (ii) proactively develop policies for both sex and labour trafficking offences, especially for minority populations, (iii) develop joint efforts with Myanmar’s neighbouring countries to monitor cross-border human trafficking within the region, (iv) engage in dialogue with its neighbouring countries regarding existing undocumented immigrants from Myanmar to its neighbouring states.

[Human trafficking and ethnic minority problems in Myanmar]

INTRODUCTION Myanmar has been criticised as one of the worst offenders for human trafficking in the world as a source and transit country. Human trafficking – forced labour and prostitution of men, women, and children – is prevalent within the country and across borders. Ethnic minorities are the most vulnerable to trafficking due to heightened conflict with the Burmese military government, as well as the marginalisation of their rights within Myanmar. Considering that ethnic minority groups account for approximately one-third of Myanmar’s population, the number of victims of human trafficking belonging to minority groups should be of great concern. Therefore, it is important to examine the relationship between a rise in human trafficking and increased minority insurgency in Myanmar. To answer this question, this paper will first examine the historical background of conflict between minority groups and the Myanmar government, the extent of human trafficking in Myanmar, and how human trafficking contributed to the militarisation of certain minority groups. Then, I will identify issues with the Myanmar government’s current policies, specifically those of Aung San Suu Kyi. Finally, I will recommend policies for Myanmar to address ethnic minority issues and to approach human trafficking problems in Myanmar more effectively.

BACKGROUND ON THE CONFLICTS BETWEEN ETHNIC MINORITIES AND THE MYANMAR GOVERNMENT Ethnic minority groups were officially brought into Burma, Myanmar’s former official name until 1989, during British colonisation from 1824 to 1948 when Britain annexed different ethnic areas within the region into one British Burma. Before

109

THE PUBLIC SPHERE | 2017 ISSUE

the British annexation, these ethnic minorities had separate and diverse histories and origins. For instance, the Kachin, Wa and Shan are also found in China; the Karen, Mon and a part of Shan come from Thailand border; the Chin and Naga from India and the Rakhine and Rohingya Muslims from Bangladesh (Smith, 1994, p.17). Each ethnic group has its own language, culture and national identity. Since British colonisation these ethnic groups have established their own minority political parties, further developing a strong sense of identity that is distinct from mainstream Burmese society, each ethnic group has already been departing away from the Burmese society. Accordingly, these minority groups have developed a strong sense of self-identity over time. FIGURE 1 ETHNOLINGUISTIC MAP OF MYANMAR

Key: Bamar = White Chin = Grey Kachin = Light green Rakine = Light blue Bamar & Mon-Khmer areas = Pink Shan = Brown Karen = Red Karen & Bamar areas = Green Mon-Khmer = Yellow Bamar & Shan areas = Blue Other = Orange Created using data from: M.J. Smith, "Burma: Insurgency and the Politics of Ethnicity" (London: Zed Books Ltd 1991)

Source: Dahgaypaw Myra, “Myanmar Population count comes with controversy,” Al-Jazeera, March 2014.

110

[Human trafficking and ethnic minority problems in Myanmar]

As one can see from the map on the previous page, Myanmar is composed of various ethnic groups that coexist throughout the country. Despite such existence of minority populations, the military government and the Tatmadaw (Burmese armed forces) have historically been dominated by the majority Burmese or Bamar tribe. The Bamar tribe also regarded other ethnic groups to be “uncivilised” (Smith, 1994, p.20). Therefore, policies that favoured the Burmese majority language and culture clashed with the strong presence of diverse ethnic groups, inevitably leading to internal conflict. In 1989, the country was renamed to Myanmar because the military government wanted to be more inclusive of ethnic groups beyond the Burmese majority. However, the name change also aroused many controversies within and outside of the country, especially since the Burmese people felt they did not have proper input into the change (Schiavenza, 2014). Ethnic minorities also felt that the new name, Myanmar, was chosen without considering any of the country’s linguistic differences. Internationally, many human rights organisations and the United States, saw the name change as a concession to the military government since the name was adopted by the repressive government. Likewise, there are many factors in Myanmar that caused internal conflict between the minority population and the Myanmar government.

GRAVITY OF CROSS BORDER HUMAN TRAFFICKING IN MYANMAR Human trafficking has become a major human rights issue in Myanmar. Not only does human trafficking within Myanmar raise a big concern, but cross border trafficking is also a critical problem. Most victims are from minority states such as Kachine and Rakhine, places that tend to suffer from political instability, growing insecurity and lack of recognition from the Myanmar

111

THE PUBLIC SPHERE | 2017 ISSUE

government (such as in the case of Rohingya). Because demand for trafficked labour is high from bordering nations such as China and Thailand, human trafficking has flourished as a robust underground business for Myanmar. FIGURE 2 TRAFFICKING ROUTES IN MYANMAR

145 290 Kms

Trafficking routes Known sources areas Known destination areas (mining) Known destination areas (fisheries)

Source: “Siren Human Trafficking Data Sheet,” United Nations Inter-Agency Project on Human Trafficking (UNIAP), March 2009.

112

[Human trafficking and ethnic minority problems in Myanmar]

Figure 2 clearly identifies human trafficking routes from Myanmar to its neighbouring states. For instance, the Dry Zone of Myanmar, which refers to the central regions of Magway, Mandalay and southern Sagiang and is home to approximately one fourth of the total population of Myanmar, is a major trafficking transit area that connects cross border trafficking routes to minority states in Myanmar. There are also many routes that traffic minority victims to Laos (UNIAP, 2015). These routes demonstrate how human trafficking in Myanmar has escalated into a regional underground operation. According to the United Nations, minority groups are trafficked to China, Malaysia and Thailand mostly for labour exploitation. For female victims, they are often subject to sexual exploitation and domestic work. Children are forced to either work as beggars or petty thieves in streets (UNIAP, 2015). In the case of human trafficking to China, victims tend to be from Kachin and Northern Shan States. Most of the victims are women, who are usually trafficked to the Yunnan Province in China as prostitutes or as “brides for sale”. These “brides” are sold to unmarried men in Chinese rural areas and are often forcibly married and subjected to domestic violence and exploitation (Blanchard, 2007). Thailand is also a major market that has high demand for trafficked labour. The main exploitative sites are in Bangkok, Samut, Mae Sai, and Mae Sot. Human traffickers supply Thailand with “slave labourers” mostly for its agricultural and fishing industries. For instance, Thai fishing industries are notorious for exploiting persecuted ethnic minorities from Myanmar. A Guardian news report estimated that at least 12,000 Rohingya migrants, an ethnic minority from the Rakhine state in Myanmar, were sold to fishing industries each month as of 2014 and were exploited as slave labourers and sex slaves (Stoakes et al., 2015). In addition, upon the agreement of the Special Economic Zone (SEZ) between Myanmar and Thailand in 2008, Thailand’s agricultural businesses were granted the rights to engage in sizeable agricul-

113

THE PUBLIC SPHERE | 2017 ISSUE

tural projects within Myanmar’s border regions that are inhabited by ethnic minorities. During these projects, however, The New York Times (2015) reports show that Thai businesses largely have utilised trafficked labour, consisting of ethnic minority groups such as the Karen. The reports also suggested that traffickers may have collided with Thai officials to smuggle ethnic minorities into Thailand. Notably, “unrecognised” ethnic groups such as the Rohingya are the main victims because they do not possesses identification documentation from the Myanmar government. Most of the Rohingya victims were asylum seekers who were transiting through Thailand to reach Malaysia, but were instead sold into forced labour in Thailand (Buckley, Berry, 2015).

VULNERABILITY OF ETHNIC MINORITY GROUPS TO HUMAN TRAFFICKING Due to the lack of ethnic minority rights, the government’s hostile policies towards minority populations, and widespread poverty, ethnic minority groups have become prime targets for human traffickers. In particular, ethnic minorities are large victims of cross border human trafficking. The United Nations Inter-Agency Project on Human Trafficking (UNIAP) reported that 73 per cent of human trafficking victims in Thailand and Malaysia are from the Shan states while 36 per cent of victims in China are from minority populations in central dry zone areas (UNIAP, 2015). Because most ethnic minorities reside in the areas bordering Thailand and China, they are frequent targets for human trafficking. Not only does proximity to the market make ethnic minority groups ideal targets for human trafficking, but also their generally lower economic status. According to the U.S. Department of State, reports indicate that Malaysian government officials purposefully lured job-seeking ethnic minorities to the Malay-

114

[Human trafficking and ethnic minority problems in Myanmar]

sia-Thailand border in order to hand them to human traffickers once they are formally deported (Ngiam, 2012). There were also several cases where ethnic minority workers were hired in neighbouring countries but were abused as slave labourers by foreign employers. Economic hardship faced by ethnic minority groups is a significant vulnerability factor, as many ethnic minority workers seek to find jobs in bordering nations such as China and Thailand that offer higher wages. Human traffickers lure minority workers by falsifying themselves as brokers (UNIAP, 2015), and then selling their victims as forced labourers or sex slaves. The Myanmar governmentâ&#x20AC;&#x2122;s policies towards ethnic minorities, especially regarding the displacement of minority groups, have significantly increased the likelihood of human traffickers targeting minority populations. The Myanmar government forcibly relocated ethnic minority groups, most prominently done after 1988 staged coup by the military government. The government justified its actions by claiming that the forced relocation was one of its counter-insurgency policies. The loss of their lands meant that ethnic minority groups were unable to sustain their livelihoods through their primary economic activity, farming. As a result, these groups were more likely to become indebted, impoverished, or homeless and thus more susceptible to human trafficking (Smith, 1994, p.71). There is also evidence that government officials have engaged in human rights abuses during this dislocation process. The United States Department of State report (2013) suggested that military personnel within Myanmar were serious perpetrators of forced labourers and child soldiers, especially in ethnic minority areas. According to this report, since the dissolution of a ceasefire with the Kachine army, conflicts with minority populations have displaced approximately 75,000 Kachine and 115,000 Rakhine residents, which left these groups highly vulnerable to forced labour and sex trafficking (U.S. Department of State, 2015). Most importantly, the Myanmar government has refused to recognise some of these displaced ethnic

115

THE PUBLIC SPHERE | 2017 ISSUE

minority groups such as the Rohingya, and the government has been accused of deliberately ignoring human rights violations and economic despair in these regions. Essentially, inhumane treatment and policies by the central government have destroyed living circumstances for some ethnic minorities, thereby causing them to be more exposed to human trafficking.

ARMED CONFLICT BETWEEN MINORITY GROUPS AND THE MYANMAR GOVERNMENT Lack of protection from the Myanmar government and the large presence of human trafficking in regions where ethnic minority groups reside has forced them to develop their own measures of protection, resulting in the formation of armed groups amongst ethnic groups in Myanmar. Although some armies, such as the Karen National Liberation Army (KNLA), were formed at the end of British colonial era, many others such as the Kachin Independence Army (KIA) were established more recently, in order to rebel against the government. These insurgencies against the military government not only sought more autonomy within their regions, but they also felt that their basic ethnic minority rights listed in the Panglong Agreement, an agreement between the Kachine leaders and General Aung San in 1947 that guarantees autonomy and internal administration of ethnic areas, were being violated. In the case of the “Asian highway” project that the Myanmar government ambitiously designed, numerous reports indicate that forced labour was used in the construction. The project also resulted in the destruction of agricultural lands in Karen state and other areas (Raj, 2008). Most importantly, there were frequent allegations where government authorities and the military took advantage of these destroyed ethnic societies to promote their human trafficking. Such cases consequently aroused general distrust and opposition

116

[Human trafficking and ethnic minority problems in Myanmar]

to the Myanmar government. Therefore, the militarisation of ethnic minority groups was predominantly caused by their mistrust of the Myanmar government. Despite such criticism, the Myanmar government has steadily made progress in stopping armed conflicts with minority armed groups. Figure 3 below shows how the Myanmar government agreed to a ceasefire with major armed groups as well as to assimilate some of these armed militias into the Myanmarâ&#x20AC;&#x2122;s Border Guard Forces. Accordingly, it seemed that the Myanmar governmentâ&#x20AC;&#x2122;s assimilation strategy was effective in ending long lasting armed conflicts with minority armed groups. FIGURE 3 THE NUMBER OF MINORITY COMBATANT GROUP THAT HAVE BEEN ABSORBED INTO THE BORDER GUARD FORCES

Source: Myanmar Peace Monitor, Oxford Analytica.

117

THE PUBLIC SPHERE | 2017 ISSUE

However, the government’s plan to assimilate armed opposition groups into Tatmadaw-commanded Border Guard Forces has intensified confrontations between the Tatmadaw and ethnic minority armed groups. Experts stress that ethnic minorities mistrust the Myanmar government and the army, arguing that there has been a “lack of genuine will by the Myanmar government to listen to the [ethnic minorities’] demands for ethnic rights and self-determination” (Sot, 2012). In parallel, as part of the “four cuts strategy” (an attempt to cut off food, funds, intelligence, and recruits to the insurgents), the Tatmadaw has perpetrated widespread violations of economic, social and cultural rights, including the deprivation of means of livelihood, excessive taxation and coercion. By these and other measures, ethnic minorities are discriminated against, or in extreme cases, not recognised as citizens of Burma (Open Democracy, 2016). These resettled ethnic minorities were “forced” to migrate to Thailand or China where they were then often used as forced labourers or, for women, used as sex slaves. Due to Tatmadaw’s disinterest in truly integrating opposition armed forces into the Burmese army, its strategies have backfired and increased resentment towards the Tatmadaw. Ethnic minorities frequently regard them as invaders and perpetrators (Smith, 1994, p.71-95). To some extent, ethnic populations began to blame the Myanmar government for economic despair and increased human trafficking in their regions, aggravating tensions between the Myanmar government and ethnic minorities.

ANALYSIS It is important to discuss the interrelationship between the Myanmar government’s policies towards minority populations and the rise of human trafficking of ethnic minorities. The most significant policy that has exacerbated the situation of human

118

[Human trafficking and ethnic minority problems in Myanmar]

trafficking has been the Burmese counter insurgency operations in ethnic minority regions. To begin with, the Myanmar government categorised various ethnic minority regions as war zones. It regarded all minority citizens as potential insurgents to justify various human rights abuses in ethnic minority regions. Government troops also forcibly seized properties, destroyed homes and killed villagers, thereby making it almost impossible for minority populations to reside in their traditional lands (Smith, 1994, p.73). The Myanmar government’s displacement policies towards ethnic minorities notably increased the vulnerability of ethnic minorities to human trafficking. For instance, the government’s displacement program forced almost 20 per cent of the Karenni and Palsung population to migrate to new lands (Smith, 1994, p.69). These minorities whose lands were seized without proper compensation from the government suffered severe economic hardship. Likewise, the Myanmar government was tolerant of their own officials committing human right abuses. As a result, human traffickers were able to take advantage of these displaced “refugees”. In this sense, the Myanmar government’s policies towards minority populations have aggravated the vulnerability of minority populations by promoting displacement and economic difficulties, in turn facilitating cross-border human trafficking. The human trafficking problem in ethnic minorities and armed conflict in Myanmar are extraordinarily intertwined. According to Martin Smith (1994), ethnic minorities are more vulnerable to trafficking when they are going through conflicts, especially armed conflicts, with the Burmese army. Armed conflicts and Tatmadaw’s policies towards ethnic minorities resulted in numerous displaced minority populations, who are inevitably targeted by human traffickers. On the other hand, the lack of systematic protection for ethnic minorities in Myanmar contributed to the formation of individual armies for ethnic minorities. After all, minorities have been looking for opportunities to gain more autonomy from the Myanmar government. The absence of ethnic

119

THE PUBLIC SPHERE â&#x20AC;&#x2020;| â&#x20AC;&#x2020;2017 ISSUE

minority rights in Myanmar and the rising presence of human traffickers have given minority groups a plausible justification to arm themselves as a form of self-protection. The Myanmar government then categorised these minority groups with armed capabilities as insurgencies and implemented military campaigns to dismantle these potential threats. In the end, increased tensions caused more economic despair and displacement of minority groups, which further increased their vulnerability to human trafficking. The interrelationship between human trafficking and armed conflict in Myanmar is an interesting case when compared to the traditional crime-conflict nexus literature. The crime-conflict nexus states that armed conflicts can create conditions for organised criminal activities. The report from the U.S. Department of State shows that human trafficking existed in Myanmar long before armed conflicts between Burmese army and minority insurgencies started. As conflict increased over time, the number of victims of human trafficking in minority regions has exponentially increased as well. For example, UNIAP reported that the number of human trafficking victims that have been rescued from the region increased from 540 in 2002 to 824 in 2005, a period during which the conflict heightened between the Myanmar government and ethnic minority groups (UNIAP, 2015). This trend could be explained by economic deterioration in ethnic minority regions due to armed conflict, which fostered cross-border trafficking to bordering nations. However, evidence has yet to show spill-over effects of human trafficking for minority insurgency groups. Some reports point out possible involvement of Myanmar and Thai officials in human trafficking, but there is no evidence that indicates a connection with minority insurgency groups. After all, human trafficking is a profitable business where there is enough demand in the market, both domestically and internationally. One possible explanation is that some of these insurgency groups were financed by other parties and financial

120

[Human trafficking and ethnic minority problems in Myanmar]

means; for instance, IHS Jane’s Intelligence Review in December 2013 reported that China is discretely supporting the Wa group both financially and militarily (Schearf, 2013). Other insurgency groups such as the Shan already have their own criminal enterprises, such as opium trade, to generate revenue. In very rare cases like the Karen, they do not need criminal enterprises to fund their insurgencies because they already possess enriched mineral resources and hydroelectric powers. Regardless of each insurgency group’s motives, there is no evidence that any are directly involved in the human trafficking business.

CONCLUSIONS FOR ETHNIC CONFLICTS IN MYANMAR For ethnic minority groups in Myanmar, human trafficking is a violation of their basic human rights. The Myanmar government’s failure to protect minorities from these criminal activities has damaged minority groups’ trusts in the government. Regardless of whether the Tatmadaw is to blame for this trafficking or not, the central government still has a duty to protect its people from such criminal activities. Perhaps minority groups have been hesitant to fully integrate into Burmese society because of this. At least minority groups who have their own military are comparatively less vulnerable to human traffickers than other minority groups without any form of armed protection. However, the problem with having militarised minority groups is that the Myanmar government is not tolerant of such military capabilities. Thus, it is likely that the government would categorise these groups as insurgencies and target them with counter-insurgency strategies. Through past examples, the Myanmar government utilised “anti-insurgency” propaganda to abuse minority rights. The Tatmadaw has displaced various minority groups from their home lands, has used its population for

121

THE PUBLIC SPHERE | 2017 ISSUE

forced labour, and has committed various human right abuses in the name of patriotism. In the end, devastated minority populations, largely by such harsh measures by the government, became more vulnerable to human trafficking. In this sense, Myanmar has fallen into an inescapable cycle when it comes to dealing with minority conflicts. Without proper measures from the central government to protect its minority groups from human trafficking, minority populations militarise on their own for the sake of self-protection. Because the Myanmar government views minority military groups as a threat, it will inevitably lead to armed conflict between those two groups.

POLICY RECOMMENDATIONS FOR MYANMAR I will recommend possible policies the government of Myanmar can adopt to solve this seemingly inescapable cycle of ethnic minority problems. To do so, I will first identify problems and issues with Myanmar’s current approaches towards minority rights. Lastly, I will present my recommendations for tackling human trafficking issues and ethnic minority problems.

Issues with Myanmar’s current policies towards minority problems The first issue is that Aung San Suu Kyi’s government is reluctant to prioritise ethnic minority issues in its agenda. When Aung San Suu Kyi successfully overtook the Myanmar governing body from the previous military regimes, many hoped that she would ameliorate the challenges facing minorities in Myanmar. Despite such hopeful expectations, Aung San Suu Kyi’s government has announced it will prioritise a peaceful democratic transition and economic development over the minority conflicts. Although Aung San Suu Kyi did not comment directly on the Rohingya re-

122

[Human trafficking and ethnic minority problems in Myanmar]

gion, she has largely showed antipathy towards the Muslim population’s concerns. U Win Htein, a spokesman and leading figure in Suu Kyi’s National League for Democracy (NLD), openly said that NLD will not prioritise Rohingya problem: “We have other priorities – peace, the peaceful transition of power, economic development and constitutional reform” (Rigby, 2015). In addition, an article in the Independent points out that “[Aung San Suu Kyi] has lamented the violence in Arakan State but has refused to endorse the positions of organisations such as Human Rights Watch, which have blamed Arakan’s Buddhists for the persecution of the Muslims” (Popham, 2016). Thus, despite Aung San Suu Kyi’s major push for liberalisation and democratisation, her agenda does not seem to include ethnic minorities in Myanmar. Another issue is that Aung San Suu Kyi’s government policies could potentially aggravate ethnic tensions in Myanmar. The Buddhist nationalists, many of whom are criticised for being extremely racist against Muslim minorities, are some of the biggest supporters of Aung San Suu Kyi. Additionally, Aung San Suu Kyi’s government did not change the previous regime’s policies towards the Rohingya population – the regime had previously officially stated that “Rohingya are interlopers and don’t deserve citizenship rights.” Furthermore, the Aung San Suu Kyi’s government refused to recognise human right violations against the Rohingya, a continuation of the policy from the previous military government (Tharoor, 2015). The government urgently needs to address human rights violations against minorities in the Rohingya region. The Telegraph estimates that “around 140,000 have been forced to live in camps for internally displaced people since violence erupted in 2012, and the remainder – many of whom have lived in Burma for decades – face major restrictions on their freedom of movement” (Rigby, 2015). In addition, the Rohingya are the most targeted subjects for human traffickers in the region. The government’s hesitation towards these issues has garnered strong criticism from the minority population. Conse-

123

THE PUBLIC SPHERE | 2017 ISSUE

quently, minorities are casting scepticism on Aung San Suu Kyi’s government and openly denying their support towards Aung San Suu Kyi. For instance, Shwe Maung, one of three Rohingya Muslim Parliamentarian, claimed “he would not be given a vote in the election” (Baulk, 2015). Despite growing disapproval from the ethnic minority population, the Myanmar government is still hesitant to comply with the ethnic minorities’ demands. The hesitation is in part because of the prevalence of Buddhism nationalism in current Myanmar politics and society. For instance, Buddhism played a significant role in rebelling against the military regime and they are the biggest supporters of Aung San Suu Kyi’s political force (Mcgowan, 2012). Considering Burmese Buddhists’ history of ethnocentrism and xenophobia of other cultures, supporting ethnic minorities groups like Rohingya could back-fire on Aung San Suu Kyi’s government. For these reasons, Aung San Suu Kyi’s government is wary of losing its main support, especially when the country is still in the middle of a political transition. However, that inaction from the Myanmar government has aggravated the tension between the Buddhist majority and other ethnic minorities. While ethnic minorities feel they are being politically marginalised, some of the Buddhist majority feel that Aung San Suu Kyi’s government has not been punitive enough on uprisings from ethnic minorities. No matter which side Aung Sung Suu Kyi’s government decides to support, it will likely to create more division in Myanmar society.

Recommendations In the following sections, I will suggest four recommendations for the Myanmar government and its neighbouring states to improve the cross border human trafficking situation for Myanmar’s ethnic minority population. 1. Develop measures to address the core grievances of

124

[Human trafficking and ethnic minority problems in Myanmar]

ethnic minoritiesâ&#x20AC;&#x2122; problems. These measures should not be constrained by traditional Buddhist nationalism, thereby allowing the Myanmar government to adopt appropriate strategies to resolve ethnic minority problems. Measures should include negotiating new treaty terms that guarantee minority groupsâ&#x20AC;&#x2122; basic rights and citizenship. Most importantly, new measures should ensure that the Myanmar government will put an end to arbitrary land confiscation, forced relocations, and oppression of minorities from expressing their opinions and beliefs. Aung San Suu Kyi has liberated ethnic Burman from previous military regimes, but other ethnic minorities have been left out of this democratisation. Aung San Suu Kyi should adopt policies that resolve issues facing ethnic minorities, especially in Rakhine State where Rohingya Muslims are the target of severe human right abuses. The international community is increasingly criticising her government for neglecting minority right issues. Human rights advocacy groups have spoken out in protest. Nicolas Kristof, a columnist from the New York Times reports that Human Rights Watch, Fortify Rights and United to End Genocide have continuously reported brutality and unhumanitarian actions against the Rohingya (Kristof, 2016). The United Nations General Assembly passed a resolution in 2013 and other statements urging the Myanmar government to grant the Rohingya full citizenship rights. Since the majority of the population in Myanmar is Buddhist, Aung San Suu Kyi understandably does not want to adopt policies that will arouse Buddhist opposition. However, if Aung San Suu Kyi fails to address minority human rights issues, her government will also face the same domestic and international legitimacy issues that the previous military regime encountered. Therefore, Aung San Suu Kyiâ&#x20AC;&#x2122;s government should address human rights abuses against ethnic minorities, despite their unpopularity with the Buddhist population for the purpose of long-term peace and legitimacy. Finally,

125

THE PUBLIC SPHERE â&#x20AC;&#x2020;| â&#x20AC;&#x2020;2017 ISSUE

the Myanmar government should legally recognise the Rohingya as a minority group in Myanmar and allow government agencies and NGOs to gather and access relevant data and resources that specifically target the Rohingya population. Since the Rohingya people are the most susceptible to cross-border human trafficking, these statistics are crucial for developing policies to protect the Rohingya. 2. Develop proactive policies for both sex and labour trafficking offenders, especially against traffickers of minority populations. Such policy options include increasing efforts to investigate and prosecute traffickers, implementing a clear victim identification process and protective measures, releasing data about human rights abuses against ethnic minorities in order to increase public awareness, allowing more authorities to participate in anti-trafficking task forces (ATTF), and granting the UN and other NGOs greater access to investigate trafficking issues. The administration can also improve coordination between the policymakers, prosecutors, and anti-trafficking task force (ATTF) to improve the system. All of the proactive policies above would help the government of Myanmar tackle the roots of the problem. There is already substantial domestic and international awareness of human trafficking in Myanmar, especially of minority groups. So far, Myanmar has developed anti-trafficking task forces, allowed NGOs limited access to investigate human trafficking issues, and attempted to develop identification measures. However, many experts such as Raj Anil, have expressed skepticism of the effectiveness of such measures (Raj, 2008). Both internal and cross-border human trafficking numbers have not decreased. The role of anti-trafficking task forces is fairly limited and NGOs are not allowed to participate in victim identification and data gathering. I recommend more proactive measures focusing on human trafficking issues

126

[Human trafficking and ethnic minority problems in Myanmar]

in order to truly address the problem. Documentation of human rights abuse cases is important because it helps policymakers to understand the magnitude of the problem as well as to adopt more effective policies for human trafficking issues. In addition, the government and NGOs should implement measures to identify vulnerable ethnic minority groups and to partner with international human rights organisations to develop more proactive protection measures. Since the governmentâ&#x20AC;&#x2122;s reactive policies towards human trafficking of ethnic minorities have proven to be ineffective, it is necessary to adopt more proactive actions. 3. Develop joint efforts with Myanmarâ&#x20AC;&#x2122;s neighbouring countries in order to mitigate cross-border human trafficking as well as protect trafficked Burmese in these nations, particularly Thailand. There are a large number of documented cases of cross-border human trafficking by NGOs, the UN, and foreign media. Myanmar is ethnically diverse and consists of ethnic groups originating from many of its bordering nations such as India, Bangladesh, China, Tibet, Laos and Thailand. Thus, it is important for the Myanmar government to realise that solely adopting domestic measures to prevent cross-border human trafficking is not enough. Instead, the Burmese government should coordinate directly with its neighbouring nations, such as Thailand, to establish a joint anti-trafficking task force (ATTF) in order to pressure human traffickers from both sides of the border. Furthermore, Myanmar should work with its neighbouring nations to grant NGOs and other international organisations unrestricted access to cross-border trafficking routes, known black markets, exploited labour sites and places with the presence of human trafficking victims. International organisations can proactively gather data and provide up-to-date statistics for Myanmar and neighbouring

127

THE PUBLIC SPHERE | 2017 ISSUE

states to evaluate the magnitude of the issue and combat trafficking. 4. Engage in dialogue with neighbouring countries to resolve the status of refugees and trafficked people from Myanmar to its neighbouring states. Myanmar also needs to address the undocumented immigrants that were already trafficked to its neighbouring states. The Bangkok Post argued that “the joint effort between [Burma and Thailand] to verify the nationality of the workers who have illegally entered Thailand failed because the Myanmar government does not care about the plight of its people” (The Bangkok Post, 2007). In fact, the Myanmar government has complicated the identification process by demanding these migrants to return to Myanmar for verification. The Myanmar government should first actively collaborate with its neighbouring nations on the migration issue. By doing so, Myanmar can be fully informed of the status and situations of migrants in its neighbouring states to develop relevant solutions to guarantee their safety and rights. Then, the government should engage in meaningful dialogue and diplomacy with its neighbouring states to protect and promote the rights of these migrant workers, the majority of whom are victims of human sex and labour trafficking.

CONCLUSION Myanmar is currently trapped in a vicious cycle, whereby ethnic minorities are marginalised, leading them to militarise, which in turn leads to government crackdown and further marginalisation. This cycle historically leaves minorities extremely vulnerable to trafficking both within Myanmar and across borders to neighbouring countries. In order to break this cycle, I recom-

128

[Human trafficking and ethnic minority problems in Myanmar]

mend that the Myanmar government should develop measures that address core grievances of ethnic minorities such as forced relocation issues and citizenship rights, develop proactive policies to stem both sex and labour trafficking offences, establish joint efforts with its neighbouring states to alleviate cross border human trafficking, and finally engage in dialogue with neighbouring states to recognise and protect existing undocumented immigrants in those states who are often victims of human trafficking. By taking these steps, the Myanmar government can start to truly establish democratic society its people have been hoping for, one that is inclusive of all ethnic groups, and sincerely combat its human trafficking problems.

REFERENCES Baulk, David. "Ethnic Minorities in Myanmar Denied Vote as Aung San Suu Kyi Claims Power." The Guardian. 2015. Accessed April 20, 2016. http://www.theguardian.com/ global-development-professionals-network/2015/nov/26/myanmar-millions-of-ethnic-minorities-denied-vote-as-aung-san-suu-kyi-claims-power-rohingya. Blanchard, Ben. "China Claims Progress Fighting Human Trafficking." Reuters. December 12, 2007. Accessed December 7, 2015. http://www.reuters.com/article/us-china-trafficking-idUSPEK11308820071212. Buckley, Chris, and Ellen Barry. "Rohingya Women Flee Violence Only to Be Sold Into Marriage." The New York Times. August 2, 2015. Accessed December 7, 2015. http:// www.nytimes.com/2015/08/03/world/asia/rohingya-women-flee-violence-only-tobe-sold-into-marriage.html?smid=fb-share&_r=1. Horsey, Richard. "Ending Forced Labour in Myanmar." Routledge Contemporary Southeast Asia Series, 2011. "Human Security in Burma: An Unreachable Conclusion?" Open Democracy. December 1, 2016. Accessed December 7, 2015. https://www.opendemocracy.net/opensecurity/ thanawat-pimoljinda/human-security-in-burma-unreachable-conclusion. Kristof, Nicholas. "Myanmar's Aung San Suu Kyi: The Dark Side of the Hero." The Sydney Morning Herald. Accessed April 20, 2016. http://www.smh.com.au/comment/myanmars-aung-san-suu-kyi-facing-ethnic-cleansing-coverup-20160127-gmf0wu.html Mcgowan, William. "Aung San Suu Kyi's Buddhism Problem." Foreign Policy Aung San Suu Kyis Buddhism Problem Comments. Accessed April 20, 2016. http://foreignpolicy. com/2012/09/17/aung-san-suu-kyis-buddhism-problem/. Ngiam, Marie. "Human Trafficking and Ethnic Minorities: The Case of Burma." March 19, 2012. Accessed December 7, 2015. http://rightnow.org.au/writing-cat/article/humantrafficking-and-the-targeting-of-ethnic-minorities-the-case-of-burma/.

129

THE PUBLIC SPHERE | 2017 ISSUE

Popham, Peter. “Aung San Suu Kyi: What the ‘interviewed by Muslim’ BBC Today Programme comment can tell us about her views.” The Independent. Accessed April 20, 2016. Raj, Anil. "Under the Iron Thumb: Forced Labour in Myanmar." Topical Research Digest: Human Rights And Contemporary Slavery, 2008. Rigby, Jennifer. "Aung San Suu Kyi Aide: Rohingya Are Not Our Priority." The Telegraph. Accessed April 20, 2016. Schearf, Daniel. "With Burma in Mind, China Quietly Supports Wa Rebels." VOA. January 25, 2013.http://www.voanews.com/a/chinese-support-for-wa-rebels-designed-tocounter-burma/1590718.html. "Seminar: Ethnic Conflict in Burma/Myanmar: From Aspirations to Solutions." Transnational Institute. 2013. Smith, Martin. Ethnic Groups in Burma: Development, Democracy and Human Rights. London: Anti-Slavery International, 1994. 17-95. Sot, Mae. "Myanmar's Ethnic Problems." IRIN. March 28, 2012. Accessed April 12, 2016. http://www.irinnews.org/report/95195/briefing-myanmar’s-ethnic-problems. Stoakes, Emanuel, Chris Kelly, and Annie Kelly. "Revealed: How the Thai Fishing Industry Trafficks, Imprisons and Enslaves." The Guardian. July 20, 2015. Accessed December 13, 2015. http://www.theguardian.com/global-development/2015/jul/20/thai-fishing-industry-implicated-enslavement-deaths-rohingya. Tharoor, Ishaan. "Why Does This Buddhist-majority Nation Hate These Muslims so Much?" Washington Post. Accessed April 20, 2016. "The Trafficking Situation in Myanmar." UNIAP. Accessed December 7, 2015. http://www. no-trafficking.org/myanmar.html. U.S. Department of State. “2013 Trafficking in Persons Report – Burma.” Accessed December 7, 2015. http://www.refworld.org/docid/51c2f3d118.html.

130

Is the UK’s proposed soft drink tax an effective way to reduce obesity? Genevieve Joy

Master of Public Administration in Public and Social Policy, Class of 2016 London School of Economics and Political Science

ABSTRACT Obesity is one of the biggest health and economic problems facing the United Kingdom. Nearly 60 per cent of British adults are overweight or obese. Obesity costs the UK an estimated £73 billion per year and this burden is growing. Among the many factors driving obesity, there is one clear culprit: excessive sugar consumption. There has been an international move towards taxing soft drinks, which contain a lot of sugar, as a way to combat obesity. In the UK’s 2016 Budget, the Chancellor introduced plans for a levy on soft drinks producers and importers to be implemented in April 2018. The Treasury estimates that it will raise £1.5 billion in the first three years, which means the tax rate will likely be 15-20 per cent. There are many ways to go about fighting obesity, including taxing other high-calorie foods, improving school nutrition and implementing bans on where sugary drinks can be sold. Using an analytical framework that outlines the preconditions necessary for a soft drinks tax to reduce obesity, this paper aims to answer the question: Is the UK’s proposed soft drinks tax an effective way to reduce obesity?

[Is the UK’s proposed soft drink tax an effective way to reduce obesity?]

INTRODUCTION The health problem of obesity and its associated costs in the UK The top risk factor for disease in the United Kingdom is entirely preventable. Poor diet can lead to type 2 diabetes, stroke and heart disease; nearly 60 per cent of British adults are overweight, and one-quarter of them are obese1 (OECD, 2014). The problem starts early and persists: Almost 20 per cent of ten-year-olds are obese (Public Health England, 2016) and 80 per cent of obese teenagers will remain so throughout their lives (HM Treasury, 2016b). Overall, the UK’s obesity rates are the second highest in Europe (OECD, 2014). Reducing this trend is paramount in the health agenda. Obesity is also an economic problem, costing the UK £73 billion per year (McKinsey, 2014). It has the second-worst economic impact on the country, below smoking and just above armed violence, war and terrorism (McKinsey, 2014). Obese adults are more likely to miss work, be less productive and suffer from depression (Marron et al., 2015). By 2040, obesity’s economic impact could jump from an already shocking 3 per cent of gross domestic product (GDP) to 11-14 per cent (McKinsey, 2014). Reducing obesity and mitigating its economic impact is complex. Yet among the many factors driving obesity, there is one clear culprit: sugar consumption. The Chief Medical Officer of the UK government has highlighted reducing sugar consumption as a priority in the fight against obesity and type 2 diabetes, which cost the National Health Service (NHS) £27 billion and £8.8 billion per year, respectively (HM Treasury, 2016b). In the UK, on average children ingest 15 per cent of their daily calories from sugar, triple the daily recommended intake of 5 per cent (Public Health England, 2016). Sugar-sweetened beverages are the number one source of sugar for these children (HM Treasury, 2016b).

1. For most adults, a body mass index (BMI) of 25-29.9 is overweight; 30-39.9 is obese; and 40+ is severely obese. BMI is determined based on weight and height. For additional information: http:// www.nhs.uk/ Conditions/Obesity/ Pages/Introduction. aspx.

133

THE PUBLIC SPHERE | 2017 ISSUE

A growing number of countries have introduced taxes on soft drinks to combat sugar consumption. Soft drinks are an easy target since they constitute a clearly-defined category, are often marketed to children, have no nutritional value and contain high levels of sugar. Over 60 public health organisations, including Public Health England and the Royal Society for Public Health, support a tax on sugary beverages (HM Treasury, 2016b), so it is no surprise that the UK government has decided to move forward with one. There are many ways to go about fighting obesity, including taxing other foods, improving school nutrition and implementing bans. This paper therefore aims to answer the question: Is the UK’s proposed soft drinks tax an effective way to reduce obesity?

Sin taxes “Sin taxes” are added to products associated with negative health outcomes, such as tobacco, alcohol and sugar. They essentially charge people for their increased burden on society through direct and indirect healthcare costs. It is estimated that healthcare costs incurred by the government for obese individuals are 30 per cent higher than for their healthy peers (Withrow and Alter, 2011), although this number may be higher (Figure 1). Sin taxes force individuals to internalise the externality their choices place on society. This is a major argument for sin taxes. Another driving factor behind sin taxes is the idea of “internalities,” costs incurred upon oneself by not fully assessing future preferences or consequences. People, particularly children, prioritise present desires more than the future (Brownell et al., 2009). Deciding what to consume, for example, they do not fully account for costs of their present actions on their future selves. Sin taxes force people with self-control issues to consider future costs by adding them to the price tag (O’Donoghue and Rabin, 2006). Sin taxes are generally considered regressive as they place a disproportionate burden on lower-income individuals. Many

134

[Is the UKâ&#x20AC;&#x2122;s proposed soft drink tax an effective way to reduce obesity?]

FIGURE 1 ASSOCIATED UK MEDICAL COSTS AS BMI INCREASES, 2012 ÂŁ per capita

1,493

1,447 1,274 1,052 805

+86%

+80% +58%

+31%

BMI <25 Band Normal weight

25-29 Overweight

30-34

35-39

>40

Obese Class 1 Obese Class 2 Obese Class 3

1. Includes primary care, general practitioner prescriptions, hospitalization, accident and emergency, and outpatient care. 2003 values taken from Tigbe et al. (2003) adjusted using 2012/2013 Federation Internationale de Medecine du Sport and Health Examination Survey data on per capita UK costs in each category.

Source: McKinsey, 2014.

critics contend that their regressive nature outweighs any potential benefits, since lower-income families end up spending a higher proportion of their income on the product, in some cases displacing spending on other items (Nestle and Rosenberg, 2015). Advocates have disputed this claim by highlighting the substantial health benefits, but this remains a point of contention.

The UK tax Proposed in the 2016 Budget, the tax is expected to be levied on soft drink producers and importers in April 2018. Drinks with under 5g of sugar per 100ml remain untaxed, drinks with 5-8g per 100ml are taxed at a certain rate, and those with more than 8g per 100ml are taxed at a higher rate (HM Treasury, 2016a). Milk-based drinks and pure fruit juices without added sugar will be exempt, as they have some nutritional value (HM Treasury,

135

THE PUBLIC SPHERE | 2017 ISSUE

2016b). The Treasury estimated that it will raise £1.5 billion in the first three years. These estimates have been used to calculate the implied tax, which is expected to be 15-20 per cent (Daneshkhu, 2016), in line with Public Health England and Parliament’s recommendations (House of Commons Health Committee, 2015). The UK joins a growing group of countries including France, Hungary and Mexico that have introduced fiscal measures to curb consumption of sugary drinks. Most of these arrived post2011, and few are implemented independent of other food-related taxes, meaning causal data linking the tax to consumption is hard to establish. While the World Health Organization (WHO) has credited soda taxes as the best way to improve diet, this is based on theory, not empirical data (Nestle and Rosenberg, 2015). Success of such taxes is therefore still open to debate. The official title of the UK tax is “Soft drinks industry levy to pay for school sport” (HM Treasury, 2016a). Despite the title’s implication that the tax is a way to raise revenue, the document primarily discusses obesity and the relationship between sugar consumption, soft drinks and childhood obesity (HM Treasury, 2016a). At the heart of the tax, therefore, is the idea that reducing soft drink consumption will reduce obesity. This paper will answer the research question by analysing six hypotheses that determine whether the tax will be effective. The next section will develop the analytical framework and hypotheses, followed by an attempt to prove the hypotheses based on theory and empirical data. The final section will discuss the findings and provide policy recommendations.

THE CONDITIONS FOR AN EFFECTIVE SOFT DRINKS TAX: FRAMEWORK OF ANALYSIS This paper drew on secondary sources, primarily journal articles, white papers and relevant newspaper articles to inform the

136

[Is the UKâ&#x20AC;&#x2122;s proposed soft drink tax an effective way to reduce obesity?]

analysis and provide data on the variables of interest. Literature was found through searches on PubMed and Google Scholar and in the reference lists of studies on soft drinks and sin taxes. The analytical framework identifies the preconditions necessary for decreasing soda consumption to reduce obesity and evaluates whether the proposed tax will be effective (see Figure 2 on next page). To be true, reduced soft drink consumption will have to translate into lower obesity, which will take years to establish â&#x20AC;&#x201C; and causality is nearly impossible to infer given the many factors that contribute to obesity. Nevertheless, alongside tracking whether there is a reduction in average BMI once the tax comes into play, additional areas can be assessed. Excess calorie consumption is a major driver of obesity and soft drinks are a major source of excess calories, so the effectiveness of the tax can be evaluated across three areas: 1. Do consumers demand fewer full-sugar soft drinks? 2. Do consumers substitute calories from soft drinks with other calories? 3. Are parallel supportive policies in place? .

Hypothesis 1: Lower soft drink consumption will reduce obesity The primary assumption behind a sin tax is that it will reduce negative health impacts â&#x20AC;&#x201C; in this case, obesity. For this to be true, sugary soft drinks must constitute a significant amount of regular caloric intake. According to Public Health England (2015), soft drinks are the top source of sugar for teenagers and children in the UK, and one of the top sources for adults (see Table 1 on next page). Individuals who consume soft drinks tend to drink one can per day, which on average contains approximately 140 calories

137

THE PUBLIC SPHERE | 2017 ISSUE

FIGURE 2 ANALYTICAL FRAMEWORK

Reduced obesity

Reduced calorie consumption

Consumers demand fewer full-sugar soft drinks Tax is passed onto consumers

Consumption is elastic

Parallel supportive policies are in place

Consumers do not substitute with other calories

Companies provide more low calorie options

Low cross-price elasticity

Switch to low calorie options

Less access to soft drinks

Health education & awareness

Social stigma

Source: Author. The bottom-level boxes are the necessary preconditions for the three main criteria. The orange indicates a direct link to the soft drinks tax, while the blue signifies essential but separate policies that do not fall within the scope of this paper. As a result, the bottom orange boxes have been reformulated into hypotheses that must be true in order for the tax to be effective. Before these can be evaluated, the premise of the tax – that lower soft drinks consumption will reduce obesity – will also be assessed as the first hypothesis.

(Public Health England, 2015). Yet they do not sufficiently account for the additional calories they drink when making food decisions later, suggesting that they think of liquid energy differently from solid. Vartanian et al.’s systematic review and meta-analysis found that over 90 per cent of the studies report higher caloric intake when soft drinks were consumed (2007). This is supported by Mourao et al. (2007), who found that because liquid forms of the same foods did not lead to the same feelings of satiety, individuals consumed between 12.4-19 per cent more

138

[Is the UK’s proposed soft drink tax an effective way to reduce obesity?]

Table 1 Contributors to sugar intake in the UK, 2008-2012 % contribution to sugar intake

Adults

Teenagers

Children 4-10 years

Children 1.5-3 years

Soft drinks

Biscuits

Buns, cakes, pastries and fruit pies

Puddings

Table sugar and preserves

Confectionery

Fruit juice

Alcoholic drinks

Breakfast cereals

Source: Public Health England.

calories when their sugar was in liquid form rather than solid. Another study found that middle schoolers increased their risk of becoming obese by 60 per cent with every extra serving of a sugary beverage per day (Ludwig et al., 2010). Studies have also directly modelled the impact of a 20 per cent tax – the recommended and expected UK levy – on caloric intake, body mass index (BMI) and health outcomes (Finkelstein, 2010; Smith et al., 2010; Briggs et al., 2013; Finkelstein et al., 2013; Sharma et al., 2014; Collins et al., 2015). Two of these studies estimated a weight loss of approximately one-third of a kilogram per person per year (Finkelstein et al., 2010; Sharma et al., 2014) while another estimated a weight loss of 1.6 pounds per year and a cumulative weight loss of 2.9 pounds over time (Finkelstein et al., 2013). Another study predicted that the tax could reduce the number of obese adults in the UK by 1.3 per cent, or 180,000 people (and overweight by 285,000) – with the highest benefits to people under 30 (Briggs et al., 2013).

139

THE PUBLIC SPHERE â&#x20AC;&#x2020;| â&#x20AC;&#x2020;2017 ISSUE

Obesity is not the only outcome variable: Malik et al. (2010) found that daily consumers of sugar-sweetened beverages had a 26 per cent greater risk of type 2 diabetes than those who had less than one serving per month, while Collins et al. (2015) found that the 20 per cent tax could lead to 2,400 fewer diabetes cases, 1,700 fewer stroke and coronary heart disease cases and 400 fewer cancer cases per year between 2010-2030 in England. Highest impact would likely be in urban areas with young populations. Other studies have found small or non-statistically significant effects on BMI or obesity as a result of taxes, but these are for much lower tax rates. Fitts and Vader (2013), for example, found no significant effect on BMI with the tax rates between 2-7.25 per cent. Similarly, Powell et al. (2009) saw no significant association between the BMI of adolescents and state soda taxes. Fletcher et al. (2015) found no evidence of weight changes for soda taxes in the United States up to 12 per cent, which was the highest rate before Berkeley, California instituted its tax in January 2015. Given the scale of the obesity problem, reducing soft drinks may seem like a drop in the bucket. Yet one can of Coca-Cola has 139 calories, or 7 per cent of the daily recommended intake for an average adult. Refraining from just one can per day would have a significant impact in the course of one year. The Department of Health (2011) called for a reduction of 100 calories per person per day as a start in the fight against obesity, so the calorie contribution of soft drinks is not negligible. All other things equal, if calories are not replaced by other drinks and foods, drinking fewer soft drinks does lead to lower weight and BMI and reduces the risk of type 2 diabetes and other obesity-related illnesses. The question of whether this happens in practice will be further explored in the analysis of cross-price elasticity in hypothesis six.

140

[Is the UK’s proposed soft drink tax an effective way to reduce obesity?]

Hypothesis 2: The tax will be passed onto consumers In order for a soft drink tax to successfully reduce consumption, it must change consumers’ habits. Before a tax can change behaviour, however, the price increase has to be passed on to the customer. The majority of the tax will be borne by whichever side is more elastic: If consumers buy fewer soft drinks rather than pay the tax, then producers will likely to reduce their margins and cover the tax themselves. However, evidence from other countries with comparable taxes suggests that the tax will be passed on consumer in full or by even more than the value of tax (The Economist, 2015; House of Commons Health Committee, 2015). In France, the tax was fully paid by the customer for carbonated soft drinks and more than 85 per cent of it was passed along to consumers for flavoured water and juice (Berardi et al., 2012). In Mexico, consumers paid 12 per cent more for soft drinks, on average, after the tax was implemented, which is even higher than the tax itself of 9-10 per cent (Marron et al., 2015). The limited experience of soft drinks taxes in other countries does not of course guarantee the same outcome in the UK. If instead producers absorb the tax and continue selling their sugary drinks at the same price, consumers will have no financial motive to change their behaviour. But it is reasonable to expect that, rather than take the financial hit of absorbing the tax, producers would be incentivised to adjust their product line toward lower-sugar drinks to avoid the higher tax brackets altogether. The success of the soft drinks tax largely depends on this hypothesis being proven true – that consumers’ demand for soft drinks is also price-elastic, as will be discussed in hypothesis three. Overall, experiences from other countries suggest that the tax will be fully or mostly passed on to the consumer. Introducing a levy at the higher level, e.g. 20 per cent, further increases chances it will be passed along by producers.

141

THE PUBLIC SPHERE | 2017 ISSUE

Hypothesis 3: Soft drink consumption is price-elastic Assuming the tax is passed on to consumers, the price elasticity of demand will largely determine the impact on consumption. The estimated ranges vary based on many factors including income and the type of beverage being taxed, and there is of course inherent uncertainty in these numbers. Most empirical data is based on taxes below what proponents suggest in order to see an impact, so the potential elasticity may be even greater than what has been seen in practice. Mexico’s tax of 9-10 per cent – one peso per litre – is at the lower end of recommendations from Public Health England and other experts (Public Health England, 2015; Boseley, 2015). A widely-accepted elasticity estimate is -0.8 to -1.0, meaning that a 10 per cent increase in price would translate roughly to an 8-10 per cent decrease in consumption (Andreyeva et al., 2010). A systematic review of studies analysing the impact of price on consumption and body weight found an elasticity of -1.2 for sugar-sweetened beverages (Powell et al., 2013). These are all in line with the recent findings for the soda tax in Mexico, which led to a 6 per cent initial decrease in sales, dropping to 12 per cent by the end of the first year (Colchero et al., 2016). For the lowest-income households, it fell by 17 per cent (Colchero et al., 2016). In some European countries, comparable taxes led to a 4 to 10 per cent fall in consumption (Charlesworth, 2016). France’s soft drinks tax, a 3 per cent levy introduced in 2012, translated into a 3 per cent decrease in sales the first year and a further 2 per cent reduction in the second year (Daneshku, 2016), although drinks with artificial sweetener were also taxed. Encouraging individuals to switch to diet soft drinks is a fundamental part of the UK government’s strategy, so an even greater drop in soft drink sales can

142

[Is the UK’s proposed soft drink tax an effective way to reduce obesity?]

be anticipated given the higher tax rate and the exemption of diet beverages. Despite the evidence that higher soft drink taxes translate to lower sales, some economists argue that the mind-set of a person with a food craving may not be accounted for in the tax system. For example, Gul and Pesendorfer (2001) have suggested that the disutility of foregoing the good might outweigh the perceived future benefits, so people will continue to purchase and consume until they are satisfied. If there is any price sensitivity, however, this may reduce the satiation point slightly by introducing additional disutility from having to pay a higher price (O’Donoghue and Rabin, 2006). Similarly, Bernheim and Rangel (2004) have explored the idea that people enter a “hot state” when confronted with a tempting or addictive product, and therefore make irrational, non-price-sensitive purchasing decisions about that good. This would mean that demand in that state is extremely inelastic and a tax would simply raise the amount of money paid without changing consumption (O’Donoghue and Rabin, 2006). If true, the tax would essentially become a revenue-raising and externality-correcting tax, rather than an internality-correcting one. Yet if the revenues go toward health-promoting activities and education, as will be the case in the UK, this may still be a successful outcome even if soft drink consumption is not reduced. It is difficult to establish causality between the tax and consumption because soft drink taxes are usually implemented with other price changes or are smaller than the 20 per cent amount recommended by many public health experts, but data still indicates that taxes reduce consumption. Most elasticity estimates based on observed behaviour are for lower tax rates, so if the tax is mostly or fully passed-through to the consumer in the UK, there should be an even greater drop in consumption.

143

THE PUBLIC SPHERE | 2017 ISSUE

Hypothesis 4: The tax structure will encourage companies to produce lower-sugar soft drinks The structure of a tax drives its differential impact. The UK levy was specifically designed to encourage producers to reduce the sugar content of their existing drinks and develop new, lower-sugar product lines (HM Treasury, 2016b). The tiered approach combined with the delayed implementation of the tax gives producers two years to adjust to the clearly-delineated tax categories. Many see this as an opportunity for companies to change their marketing strategies and refocus their product development (Marron, 2016), although the industry itself is frustrated by the negative light the tax shines on its products and claim that the tax is unnecessary since the industry is already moving away from full sugar. Director of the British Soft Drinks Association (BSDA) Gavin Partington remarked: We are extremely disappointed by the Government’s decision to hit the only category in the food and drink sector which has consistently reduced sugar intake in recent years – down 13.6 per cent since 2012. We are the only category with an ambitious plan for the years ahead – in 2015 we agreed [to] a calorie reduction goal of 20 per cent by 2020. By contrast, sugar and calorie intake from all other major take home food categories is increasing – which makes the targeting of soft drinks simply absurd. (BSDA, 2016)

Existing industry trends do point to a shift toward a larger product range with lower-sugar, lower-calorie options. The top soft drink companies have been taking steps to change their formulas by reducing the sugar content and using artificial or natural sweeteners like stevia (Euromonitor International, 2015). The Coca-Cola Company almost doubled its product offering from 400 in 2004 to 700 in 2015 (Sanger-Katz, 2015), and has reduced the average calories of its carbonated beverages by 5.3 per cent since 2014. It has also invested £15 million to reduce the calories and sugar in its top drinks between

144

[Is the UK’s proposed soft drink tax an effective way to reduce obesity?]

2012 and 2014 (Coca-Cola, 2015). This push is reflected by retailers as well: In 2015, Tesco committed to dropping the sugar content of its branded drinks by 5 per cent per year beginning in 2016 (Boseley, 2015). These efforts will prove to be key in reducing consumption of full-sugar soft drinks. In 2005, the UK had a push to reduce salt consumption, which succeeded primarily thanks to producers lowering salt content, not consumers changing their purchasing patterns (Marron, 2016). The government needs the industry to continue reducing sugar content in order to wean the nation off of high-sugar drinks. It is unclear if the additional pressure of a tax is necessary to motivate that change, or if it unnecessarily alienates industry by casting them as the “bad guy”. Either way, recent trends suggest that producers will continue to shift their product range toward lower sugar and calorie drinks.

Hypothesis 5: The tax structure will push consumers toward lower-sugar beverages The UK government hopes that consumers will shift toward lower sugar beverages due to the price incentives, increased awareness around the unhealthily high sugar levels in regular soft drinks, and expanded low-sugar offerings from producers. The proposal accomplishes this goal first through taxing only sugary soft drinks – and not those with artificial sweeteners – and second through the tiered structure of the tax, which provides a financial incentive to move toward lower-sugar options. A number of countries including Finland, France and Hungary tax sugary soft drinks but also tax diet soda or bottled water, obviating a reason for consumers to avoid sugary ones in particular (Nestle and Rosenberg, 2015). By excluding artificially-sweetened drinks from the tax, UK customers have a more affordable option available. With 59 per cent of soft drinks low or no-calorie and new

145

THE PUBLIC SPHERE | 2017 ISSUE

incentives for producers to expand this offering, there is plenty of scope to move away from higher-sugar choices (BSDA, 2015). The three-tiered structure of the UK’s proposal is innovative and addresses concerns that soft drinks taxes tend to treat all sugary beverages equally. Finland and St Helena both have a sugar threshold before the tax is levied, but only Chile has a comparable tiered system, with different rates on sugary drinks with more and less than 6.25g per 100 ml. In the UK, the levy will be an estimated 24p per litre for beverages with more than 8g of added sugar per 100 ml, 18p per litre for beverages from 5-8g of sugar per 100 ml, and no tax below 5g (Institute for Fiscal Studies, 2016). The sugar content of the top soft drinks and where they fit into the tax bands can be seen in Figure 3. The UK’s tax also avoids a pitfall that other countries experienced by targeting sugar content rather than drink volume. Whereas most countries charge a tax per litre, treating the sugar of low-sweetened iced tea and Coca-Cola the same, the UK’s tax per gram of sugar ensures that a price-sensitive consumer will be pushed toward the lower-sugar options. Applying the levy as an excise tax rather than a value-added tax (VAT) ensures that consumers are fully aware of the price increase when making purchasing decisions because it is already on the sticker (Marron et al., 2015). When VAT is added at the cash register, some may only realise the price difference when they have already decided to buy (Brownell et al., 2009). A VAT would potentially benefit budget brands over lower-sugar beverages since it is proportional to the price of the product; the excise tax, on the other hand, applies the same tax rate to the same sugar content whether it is Coca-Cola or Tesco. The added benefit of the excise tax combined with the tiered system is that brand-loyal customers will be incentivised to follow their brand to a lower-sugar product rather than switch to a non-branded company. If consumers are indeed price-sensitive, then there is a clear motivation to avoid the tax. Not everyone will switch from a full-sugar

146

[Is the UKâ&#x20AC;&#x2122;s proposed soft drink tax an effective way to reduce obesity?]

FIGURE 3 SUGAR CONTENT OF LEADING SOFT DRINK PRODUCERS Red Bull Pepsi 7 up Sainsburys Classic Cola Relentless Coca-cola Tesco Tropical Juice Drink Tesco Cola Lucozade energy Sainsburys Cola Dr. Pepper Fanta Lipton Ice Tea Lemon Sprite Schweppes Lemonade Tango Lucozade Sport Tesco Lemonade

5 6 7 8 Sugar per 100ml (g)

Source: Harper et al., 2016.

soda to no-sugar beverages, though, so creating a tiered system encourages an adjustment to lower-sugar beverages. By not taxing diet soft drinks, there is a better chance that the replacement drink will be low-calorie for price-elastic individuals. If consumers are receptive to these options, it appears that the UKâ&#x20AC;&#x2122;s proposed tiered levy will indeed push consumers toward them.

Hypothesis 6: There is low cross-price elasticity between soft drinks and other high-calorie products Opponents of soft drink taxes contend that people will simply offset calories from the taxed beverage, such as by replacing soft drinks with untaxed high-calorie beverages like fruit juice, which is naturally high in sugar, or milk-based drinks, or if they increase their consumption of sugary foods. There is not enough data and

147

THE PUBLIC SPHERE â&#x20AC;&#x2020;| â&#x20AC;&#x2020;2017 ISSUE

too many confounding factors to conclude how the public will react when the tax is implemented, but empirical studies on cross-price elasticity and the metabolic impact of soft drinks can help predict the outcome. Studies indicate that there will be some form of substitution by other beverages or foods, but it is difficult to say how much. Brownell et al. (2009) suggest that 25 per cent of calories from sugar-sweetened beverages will be substituted by other drinks and foods. Both Barquera et al. (2008) and Colchero et al. (2015) found that higher-priced soft drinks were correlated with higher expenditure on water and milk and lower expenditure on other sugar-sweetened beverages like fruit juice and energy drinks. Dharmasena and Capps (2009) found that a 1 per cent increase in soft drink prices would see a 1.9 per cent fall in soft drink consumption and a 0.49 per cent reduction in diet soft drinks, while there would be a 0.29 per cent increase in low-fat milk, a 0.49 per cent increase in high-fat milk and a 1.15 per cent increase in fruit juice. In an analysis of the United States, individuals in states with higher soda taxes reduced their consumption of soda by 6 calories per day for each 1 cent increase in the tax; but completely replaced the calories by drinking other high-calorie beverages like milk and juice (Fletcher et al., 2010). Much more research is required on the topic to understand the effects of a larger tax and consumer behaviour, but it is clear that caloric substitution is a critical component and does exist to some degree. Research on the effects of liquid versus solid carbohydrates increasingly shows that liquids do not produce the same feeling of satiation as solids with the same calories (Pan and Hu, 2011; Malik et al., 2006) and people who drink beverages with a lot of calories do not compensate by reducing their food consumption at subsequent meals, leading to higher overall energy intake (Mourao et al., 2007; Vartanian et al., 2007; DiMeglio and Mattes, 2000). In addition, soft drinks may be complementary with other unhealthy foods like pizza and hamburgers, and their consumption may be negatively correlated with healthy eating habits overall (Vartanian et al., 2007; Fin-

148

[Is the UK’s proposed soft drink tax an effective way to reduce obesity?]

kelstein et al., 2013). Based on this line of research, even if people switch from drinking soft drinks to eating snacks, the overall calorie consumption might still go down. Substitution of regular soft drinks with diet drinks or water is expected to lower caloric consumption, but this may not necessarily be the case. In their review of a soft drinks tax in the UK, Tiffin et al. (2014) found that if diet drinks are not taxed, then a shift to those beverages would enhance the effectiveness of the tax by reducing caloric intake. Indeed, the soft drinks tax in Mexico resulted in a 4 per cent increase in bottled water sales in the first year, while sales of soda dropped (Colchero et al., 2016). Yet there are concerns that drinking diet beverages may give people a sense that they are “justified” in consuming more calories later or may increase preference for sweet foods and drinks (Mattes and Popkin, 2009). It is interesting to note that overweight and healthy-weight Americans drink the same amount of regular soda, whereas overweight Americans are more likely to drink diet soda (Mendes, 2013). It is unrealistic to imagine that a consumer will be diverted from soft drinks but will not find another product to fill the gap. If that product is diet soda, then all else being equal, calorie consumption will go down – and, to a small extent, so will obesity. But if the replacement product has calories, or if the diet drink makes the consumer feel entitled to compensate with more food, then the effect of the tax will be attenuated if not negated. The outcome of this hypothesis, which is central to the success of the tax, will depend on a combination of consumer behaviour and the effects of complementary interventions such as health education and marketing. Only a comprehensive and consistent approach will translate into an effective tax.

149

THE PUBLIC SPHERE | 2017 ISSUE

CONCLUSION Obesity is a complex problem, and the solutions are more complex still. Yet the importance of reducing it is undeniable. Waiting for a panacea will just allow the problem to grow. None of the hypotheses presented in this paper can be irrefutably proven, but theoretical and empirical data fall on the side of the tax in nearly every case: �

Evidence indicates that lower soft drink consumption will indeed reduce obesity, even if only in part;

�

The experiences of other countries tell us that the tax will be passed onto consumers, most likely in full – and since the UK tax will probably be one of the highest to date, even a partial pass-through should see results;

�

With a higher sticker price, at least 8-10 per cent of existing soft drink purchases will likely no longer be made;

�

A tiered tax structure will encourage companies to continue to produce lower-sugar soft drinks in order to avoid the tax, providing more choice to consumers;

�

Through a combination of fiscal incentives, more lowsugar and sugar-free options, and the marketing around the tax, consumers are likely to shift consumption away from sugar-added beverages.

The most uncertain hypothesis, and one central to the effectiveness of the tax, is cross-price elasticity. It is very sensitive to geography, taxes on other products, pre-existing food preferences, education and other variables. This is where parallel support-

150

[Is the UK’s proposed soft drink tax an effective way to reduce obesity?]

ive policies such as reduced access to soft drinks and increased access to water in schools, expanded nutritional education, and clarity around where the tax revenues go become essential. Moving forward with a soft drinks tax as part of a larger package of policy changes makes sense, but other policies and government initiatives should be in place to maximise effectiveness. One of the most important efforts in the soft drinks tax campaign will be addressing the argument that it is regressive. As the House of Commons Health Committee pointed out in its report on childhood obesity, “there is a strong case that [lower income families] should also derive the most benefit” from the tax (House of Commons Health Committee, 2015). The regressivity of the tax could be offset if the money is targeted toward lower-income communities, while the health benefits might also outweigh the financial burden in the long run. The UK has committed to using the tax revenue to support longer school days for secondary schools to facilitate after-school activities, increase funding for school sports, and develop the school breakfast club programme to provide healthy breakfasts to more children (HM Treasury, 2016b). As the revenue comes in, the government should be vocal and clear about where the money is going and how children are benefiting. Soft drinks are high in calories and low in nutritional value. Excessive consumption of them is associated with higher BMI, type 2 diabetes, tooth decay and consumption of other unhealthy foods. Taxing soft drinks will not be the main way of reducing obesity, as no one product is responsible for the majority of caloric intake. However, if one product category is to be targeted, and there is a strong value placed on changing young people’s habits from an earlier age, then it is a good candidate for intervention. As part of a full package of comprehensive and consistent policies, the proposed soft drinks levy will be effective and beneficial to the health of UK citizens.

151

THE PUBLIC SPHERE | 2017 ISSUE

REFERENCES Andreyeva, T., Long, M. and Brownell, K. (2010). The Impact of Food Prices on Consumption: A Systematic Review of Research on the Price Elasticity of Demand for Food. Am J Public Health, 100(2), pp.216-222. Berardi, N., Sevestre, P., Tepaut, M. and Vigneron, A. (2012). The Impact of a ‘Soda Tax’ on Prices: Evidence from French Micro Data. SSRN Electronic Journal. Bernheim, B. and Rangel, A. (2004). Addiction and Cue-Triggered Decision Processes. American Economic Review, 94(5), pp.1558-1590. Boseley, S. (2015). Mexican soda tax cuts sales of sugary soft drinks by 6% in first year. The Guardian. [online] Available at: http://www.theguardian.com/world/2015/jun/18/ mexican-soda-tax-cuts-sales-first-year [Accessed 27 Mar. 2016]. Briggs, A., Mytton, O., Kehlbacher, A., Tiffin, R., Rayner, M. and Scarborough, P. (2013). Overall and income specific effect on prevalence of overweight and obesity of 20% sugar sweetened drink tax in UK: econometric and comparative risk assessment modelling study. BMJ, 347, pp.f6189-f6189. British Soft Drinks Association, (2015). Changing Tastes: The UK Soft Drinks Annual Report 2015. [online] London. Available at: http://www.britishsoftdrinks.com/write/ MediaUploads/Publications/BSDA_Annual_Report_2015.pdf [Accessed 23 Feb. 2016]. British Soft Drinks Association, (2016). Response to the proposed Soft Drinks Levy. [online] Available at: http://www.britishsoftdrinks.com/Press-releases-/response-tothe-proposed-soft-drinks-levy [Accessed 5 Apr. 2016]. Brownell, K., Farley, T., Willett, W., Popkin, B., Chaloupka, F., Thompson, J. and Ludwig, D. (2009). The Public Health and Economic Benefits of Taxing Sugar-Sweetened Beverages. New England Journal of Medicine, 361(16), pp.1599-1605. Charlesworth, A. (2016). Nudge, nudge! How the sugar tax will help British diets. Financial Times. [online] Available at: http://www.ft.com/cms/s/0/edf07088-ecfe-11e5-bb792303682345c8.html?siteedition=uk#axzz445yazog0 [Accessed 27 Mar. 2016]. Coca-Cola. (2015). Tackling Obesity: Choice and Information. [online] Available at: http:// www.coca-cola.co.uk/stories/tackling-obesity-choice-and-information/ [Accessed 23 Mar. 2016]. Colchero, M., Popkin, B., Rivera, J. and Ng, S. (2016). Beverage purchases from stores in Mexico under the excise tax on sugar sweetened beverages: observational study. BMJ, p.h6704. Colchero, M., Salgado, J., Unar-Munguía, M., Hernández-Ávila, M. and Rivera-Dommarco, J. (2015). Price elasticity of the demand for sugar sweetened beverages and soft drinks in Mexico. Economics & Human Biology, 19, pp.129-137. Collins, B., Capewell, S., O’Flaherty, M., Timpson, H., Razzaq, A., Cheater, S., Ireland, R. and Bromley, H. (2015). Modelling the Health Impact of an English Sugary Drinks Duty at National and Local Levels. PLOS ONE, 10(6), p.e0130770. Cramer, S. (2016). Food should be labelled with the exercise needed to expend its calories. BMJ, [online] p.i1856. Available at: http://www.bmj.com/content/353/bmj.i1856 [Accessed 25 Apr. 2016]. Credit Suisse, (2013). Is Sugar Turning the Economy Sour?. [online] Available at: https:// www.credit-suisse.com/us/en/news-and-expertise/economy/articles/news-and-expertise/2013/09/en/is-sugar-turning-the-economy-sour.html [Accessed 27 Feb. 2016].

152

[Is the UKâ&#x20AC;&#x2122;s proposed soft drink tax an effective way to reduce obesity?]

Daneshkhu, S. (2016). Soft drinks groups digest UK sugar tax. Financial Times. [online] Available at: http://www.ft.com/cms/s/0/8521da6e-ec5a-11e5-bb79-2303682345c8. html?siteedition=uk#axzz43eYVTRbb [Accessed 27 Mar. 2016]. Department of Health, (2011). Healthy Lives, Healthy People: a call to action on obesity in England. London. Dharmasena, S. and Capps, O. (2009). Demand Interrelationships of At-Home Nonalcoholic Beverage Consumption in the United States. [online] Available at: http:// ageconsearch.umn.edu/bitstream/49443/2/Dharmasena%20and%20Capps%20 AAEA%20Paper%20(Milwaukee%20Wisconsin%20July%202009)%20(May1,%20 2009).pdf [Accessed 23 Mar. 2016]. Euromonitor International, (2015). Soft Drinks in Mexico. Passport. Falbe, J., Rojas, N., Grummon, A. and Madsen, K. (2015). Higher Retail Prices of Sugar-Sweetened Beverages 3 Months After Implementation of an Excise Tax in Berkeley, California. Am J Public Health, 105(11), pp.2194-2201. Finkelstein, E. (2010). Impact of Targeted Beverage Taxes on Higher- and Lower-Income Households. Arch Intern Med, 170(22), p.2028. Finkelstein, E., Zhen, C., Bilger, M., Nonnemaker, J., Farooqui, A. and Todd, J. (2013). Implications of a sugar-sweetened beverage (SSB) tax when substitutions to non-beverage items are considered. Journal of Health Economics, 32(1), pp.219-239. Fletcher, J., Frisvold, D. and Tefft, N. (2010). The effects of soft drink taxes on child and adolescent consumption and weight outcomes. Journal of Public Economics, 94(1112), pp.967-974. Fletcher, J., Frisvold, D. and Tefft, N. (2015). Non-Linear Effects of Soda Taxes on Consumption and Weight Outcomes. Health Econ., 24(5), pp.566-582. Gul, F. and Pesendorfer, W. (2001). Temptation and Self-Control. Econometrica, 69(6), pp.1403-1435. Harper, H., Ravenscroft, L. and Service, O. (2016). Sugar tax: how will it affect behaviour?. The Behavioural Insights Team. [online] Available at: http://www.behaviouralinsights.co.uk/health/behaviour-change-and-the-new-sugar-tax/ [Accessed 25 Mar. 2016]. Health Metrics and Evaluation. (2010). GBD Profile: United Kingdom. [online] Available at: https://www.healthdata.org/sites/default/files/files/country_profiles/GBD/ihme_ gbd_count ry_report_united_kingdom.pdf [Accessed 25 Apr. 2016]. HM Treasury, (2016a). Budget 2016. London. HM Treasury, (2016b). The soft drinks industry levy. [online] Available at: http://www. foodpolitics.com/wp-content/uploads/160315-Soft-drinks-stakeholder-infosheet-FINAL-2.pdf [Accessed 5 Apr. 2016]. House of Commons Health Committee, (2015). Childhood obesity â&#x20AC;&#x201C; brave and bold action: First Report of Session 2015-16. London: The Stationery Office Limited. Institute for Fiscal Studies, (2016). The Soft Drinks Levy. Knapton, S. (2015). Sugar tax to pay for treating obesity. The Telegraph. [online] Available at: http://www.telegraph.co.uk/news/health/news/11622500/Sugar-tax-to-pay-fortreating-obesity.html [Accessed 27 Mar. 2016]. Malik, V., Popkin, B., Bray, G., Despres, J., Willett, W. and Hu, F. (2010). Sugar-Sweetened Beverages and Risk of Metabolic Syndrome and Type 2 Diabetes: A meta-analysis. Diabetes Care, 33(11), pp.2477-2483. Malik, V., Schulze, M. and Hu, F. (2006). Intake of sugar-sweetened beverages and weight gain: a systematic review. American Journal of Clinical Nutrition, [online] 84(2),

153

THE PUBLIC SPHERE | 2017 ISSUE

pp.274-288. Available at: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3210834/ [Accessed 21 Feb. 2016]. Marron, D. (2016). Britain Builds a Better Soda Tax. [Blog] Tax Policy Center. Available at: http://www.taxpolicycenter.org/taxvox/britain-builds-better-soda-tax [Accessed 5 Apr. 2016]. Marron, D., Gearing, M. and Iselin, J. (2015). Should we tax unhealthy food and drinks?. Urban-Brookings Tax Policy Center. [online] Available at: http://www.urban.org/sites/ default/files/alfresco/publication-pdfs/2000553-Should-We-Tax-Unhealthy-Foodsand-Drinks.pdf [Accessed 24 Feb. 2016]. Mattes, R. and Popkin, B. (2009). Nonnutritive sweetener consumption in humans: effects on appetite and food intake and their putative mechanisms. American Journal of Clinical Nutrition, 89(1), pp.1-14. McKinsey, (2014). Overcoming obesity: An initial economic analysis. Discussion Paper. [online] Available at: http://www.mckinsey.com/industries/healthcare-systems-andservices/our-insights/how-the-world-could-better-fight-obesity [Accessed 24 Feb. 2016]. Mendes, E. (2013). Regular Soda Popular with Young, Nonwhite, Low-Income. [online] Gallup. Available at: http://www.gallup.com/poll/163997/regular-soda-popular-young-nonwhite-low-income.aspx?g_source=soda&g_medium=search&g_campaign=tiles [Accessed 5 Apr. 2016]. Mourao, D.M., Bressan, J., Campbell, W.W. and Mattes, R.D. (2007) Effects of food form on appetite and energy intake in lean and obese young adults. International Journal of Obesity, 31(11), pp.1688-95. Nestle, M. and Rosenberg, T. (2015). The whole world is watching. Soda wars. Sugar tax. US, Mexico. World Nutrition, [online] 6(11-12), pp.811-832. Available at: http://wphna. org/wp-content/uploads/2015/11/WN-2015-06-11-12-811-832-Marion-Nestle-TinaRosenberg-Soda-wars.pdf [Accessed 1 Dec. 2015]. O’Donoghue, T. and Rabin, M. (2006). Optimal sin taxes. Journal of Public Economics, 90(10-11), pp.1825-1849. OECD, (2014). Key Facts: England, Update 2014. Obesity and the Economics of Prevention: Fit Not Fat. [online] Available at: http://www.oecd.org/unitedkingdom/Obesity-Update-2014-ENGLAND.pdf [Accessed 7 Feb. 2016]. Pan, A. and Hu, F. (2011). Effects of carbohydrates on satiety: differences between liquid and solid food. Current Opinion in Clinical Nutrition and Metabolic Care, 14(4), pp.385-390. Powell, L., Chriqui, J. and Chaloupka, F. (2009). Associations between State-level Soda Taxes and Adolescent Body Mass Index. Journal of Adolescent Health, 45(3), pp.S57-S63. Powell, L., Chriqui, J., Khan, T., Wada, R. and Chaloupka, F. (2013). Assessing the potential effectiveness of food and beverage taxes and subsidies for improving public health: a systematic review of prices, demand and body weight outcomes. Obes Rev, 14(2), pp.110-128. Public Health England, (2014). Sugar Reduction: Responding to the Challenge. [online] Available at: https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/324043/Sugar_Reduction_Responding_to_the_Challenge_26_June. pdf [Accessed 6 Mar. 2016]. Public Health England, (2015). Sugar Reduction: The evidence for action. [online] Available at: https://www.gov.uk/government/uploads/system/uploads/attachment_

154

[Is the UK’s proposed soft drink tax an effective way to reduce obesity?]

data/file/470179/Sugar_reduction_The_evidence_for_action.pdf [Accessed 6 Mar. 2016]. Public Health England. (2016). Child Obesity: Obesity Knowledge and Intelligence team. [online] Available at: http://www.noo.org.uk/NOO_about_obesity/child_obesity [Accessed 7 Apr. 2016]. Sanger-Katz, M. (2015). The Decline of ‘Big Soda’. The New York Times. [online] Available at: http://www.nytimes.com/2015/10/04/upshot/soda-industry-struggles-as-consumer-tastes-change.html?_r=0 [Accessed 5 Apr. 2016]. Sanger-Katz, M. (2016). Making a Soda Tax More Politically Palatable. The New York Times. [online] Available at: http://www.nytimes.com/2016/04/04/upshot/making-a-soda-tax-more-politically-palatable.html?ribbon-ad-idx=3&rref=upshot&module=Ribbon&version=context&region=Header&action=click&contentCollection=The%20Upshot&pgtype=article [Accessed 5 Apr. 2016]. Shakespeare, S. (2016). Sugar tax leaves a sour taste for fizzy drinks brands. [online] YouGov. Available at: http://www.brandindex.com/article/sugar-tax-leaves-sourtaste-fizzy-drinks-brands [Accessed 24 Apr. 2016]. Sharma, A., Hauck, K., Hollingsworth, B. and Siciliani, L. (2014). The effects of taxing sugar-sweetened beverages across different income groups. Health Econ., 23(9), pp.1159-1184. Singh, G., Micha, R., Khatibzadeh, S., Lim, S., Ezzati, M. and Mozaffarian, D. (2015). Estimated Global, Regional, and National Disease Burdens Related to Sugar-Sweetened Beverage Consumption in 2010. Clinical Perspective. Circulation, 132(8), pp.639-666. The Economist, (2015). Stopping slurping. The Economist. [online] Available at: http:// www.economist.com/news/finance-and-economics/21679259-taxes-fizzy-drinksseem-work-intended-stopping-slurping [Accessed 7 Apr. 2016]. Tiffin, R., Kehlbacher, A. and Salois, M. (2014). The Effects of a Soft Drink Tax in the UK. Health Econ., 24(5), pp.583-600. Vartanian, L., Schwartz, M. and Brownell, K. (2007). Effects of Soft Drink Consumption on Nutrition and Health: A Systematic Review and Meta-Analysis. Am J Public Health, 97(4), pp.667-675. Withrow, D. and Alter, D. (2011). The economic burden of obesity worldwide: a systematic review of the direct costs of obesity. Obesity Reviews, 12(2), pp.131-141.

155

Are Malawi, Rwanda, Uganda and Zambia prepared to meet the Sustainable Development Goalsâ&#x20AC;&#x2122; health targets? A policy analysis Natasha Mbabazi

MSc Health Policy, Planning and Financing , Class of 2016 London School of Economics and Political Science, and London School of Hygiene and Tropical Medicine

ÂŠSarah Ju

ABSTRACT The primary aim of this paper is to evaluate and compare the health policies of Malawi, Rwanda, Uganda and Zambia in their preparedness to achieve the thirteen health targets under Goal 3, â&#x20AC;&#x153;Ensure healthy lives and promote well-being for all at all agesâ&#x20AC;? of the United Nations Sustainable Development Goals (SDGs). This project evaluates the congruency of national policies and strategies with the SDG health targets. This body of work requires secondary research and literary review of national policy documents and United Nations policy language. Knowledge management databases on ministerial websites are also used. Additionally, some informational requests for policy documents are made to ministerial staff. Malawi, Rwanda, Uganda and Zambia score similarly on sexual reproductive health integration, universal health coverage, and human resources policy language. However, in terms of policy preparedness, Rwanda ranks best on the matrices. All countries are strongest in knowledge management, with the notable exception of Malawi. The countries collectively struggled most with cross-sectoral and/or intersectional policy language. Rwandaâ&#x20AC;&#x2122;s policy are most prepared to achieve the SDG health targets. All policy recommendations will require capacity, resources and above all, political will.

[Malawi, Rwanda, Uganda and Zambia - Sustainable Development Goals’ health targets]

BACKGROUND According to the World Health Organisation (WHO), national health policies “define a country’s vision, priorities, budgetary decisions and course of action”(WHO, 2016). Policies act as a legitimate, recorded consensus of a nation’s policy makers on how to move forward in a particular area. They can serve as a point of reference on how to continue moving forward with programs and plans when there are many changes outside of the health sector such as economic crises, coups d’état, natural disasters and armed conflicts. For low-income countries, policies help donors and other stakeholders understand not only the state’s priorities, but previous accomplishments, and the unfulfilled needs of its various constituencies. Policies that are both published and easily accessible can also be used as a road map for nations with similar dynamics and demographics looking for guidance on how to proceed in executing health programs. The findings in this paper offer stakeholders a high-level perspective of strengths and gaps in policies related to the SDG health targets. From that vantage point, stakeholders have a better understanding of issue areas that need political will (the intention to carry through a policy), advocacy efforts, or further systems strengthening. Use of the matrices are transferrable to any country setting and can supplement other SDG evaluations that have begun to emerge. These four countries were analysed due to the author’s professional experience and contacts in Malawi, Rwanda, Uganda and Zambia. All four are members of the Common Market for East and Southern Africa, a free trade area of twenty member states in sub-Saharan Africa. Uganda and Rwanda are neighbouring countries located in the east and are also members of the East African Community, a regional intergovernmental organization. Similarly, Malawi and Zambia are neighbouring countries in southern Africa and members of the Southern African De-

159

THE PUBLIC SPHERE | 2017 ISSUE

velopment Community. Further, Zambia and Malawi have been relatively peaceful nations while Rwanda and Uganda have recent histories of internal strife. Armed conflicts and displaced populations have grave implications for a country’s health indicators, available human resources, and sustainability of its health systems. This study may be of interest to those comparing national health policies and service readiness within and among these regional blocs. This research questions how prepared Malawi, Rwanda, Uganda and Zambia are to achieve the thirteen health targets under Goal 3 of the UN SDGs. It will examine how well their policies measure against a preparedness framework, as well as how prepared they are to achieve each individual target. It concludes with a comparative analysis of all the countries.

DEFINING POLICY PREPAREDNESS The following five points are the indicators in which policies will be evaluated against, followed by brief justifications:

1. Policies address WHO health system building blocks The WHO is the United Nations’ specialised agency focused on health. Its mandate is to support countries in achieving their health objectives (World Health Organization, 2016). The WHO building blocks represent the core pillars of functioning and thriving health systems. It is important that policies aim to address at least one of the building blocks (WHO Western Pacific Region, 2016): Service delivery Health workforce Health information systems Financing

160

[Malawi, Rwanda, Uganda and Zambia - Sustainable Development Goals’ health targets]

Leadership/governance Access to essential medicines

2. Targets explicitly stated in policy language At what level is national policy language congruent to the SDGs target language? The SDG language is adaptable and it is widely accepted best practice that each country draft national policies in a manner that best suits their unique context. With that said, a clear correlation to the SDG language may mean a better chance of local policy translating into activities directly in line with achieving the SDG goals.

3. Strong knowledge management Policies are easily accessible to stakeholders. Jeffrey Sachs, an economist and architect of the MDGs, observed that the data available to monitor MDG progress was often outdated, if accessible at all. Making better use of technology to manage information helps strengthen programs in real-time management, advocacy and feedback (Sachs, 2012). Additionally, stakeholders (including ministerial staff) develop their work plans in a more timely and efficient manner when they are afforded easy access to policies and data. This requires countries to take advantage of increased technological connectivity.

4. Cross-sectoral and/or intersectional component Evidence of cross sectoral/ministerial collaboration. Health issues such as pollution and road safety do not typically fall under jurisdiction of health ministries. For these, cross-sectoral cooperation is needed to bring together different skillsets and expertise to achieve public health outcomes.

161

THE PUBLIC SPHERE | 2017 ISSUE

Youth and/or gender intersectionalities also determine how one will access health care, one’s ability to access it, and which health issues are priorities. Gender is defined by the WHO as “socially constructed roles, behaviours, activities, and attributes that a given society considers appropriate for men and women.” The WHO continues: “Gender norms, roles and relations influence people’s susceptibility to different health conditions and diseases and affect their enjoyment of good mental, physical health and wellbeing. They also have a bearing on people’s access to and uptake of health services and on the health outcomes they experience throughout the life-course”(WHO, 2016). The definition of youth can vary country to country but generally captures the age range of 15 to 24 (WHO, 2016). Like gender, age can determine cultural expectations in a society. For example, youths in Malawi have had trouble accessing reproductive health services from facilities operated by staff who believe youths should not be sexually active (Kambalame, 2013). Further, gender and youth can intersect in a way that needs to be acknowledged by policy makers. In Zambia, adolescent girls have higher rates of child marriage due to the vulnerability brought by their gender, age, and economic deprivation (Imbuwa, 2015).

5. Integration into work plans A national work plan exists for the policy. Work plans are a detailed account of how a policy will be accomplished. The existence of a work plan demonstrates a commitment to follow through on policies. It also means a nation is one step closer to achieving the policy’s outcomes. In Malawi, for example, some policies take more than a year from final draft form to official publication. The Government of Rwanda provides insight to the prolonged policy making process: “prior to the approval of any policy, the Government of Rwanda conducts

162

[Malawi, Rwanda, Uganda and Zambia - Sustainable Development Goals’ health targets]

extensive stakeholders’ consultation processes in accordance with the Cabinet manual. The consultation process ensure[s] that the policy adequately identify all relevant issues and proposes suitable actions to address them. At the end of a consultation process, the policy is considered as a final draft paper and submitted to cabinet for approval”(Myict.gov.rw, 2016). In essence, the sooner a policy is in plan form, the sooner execution can be expected.

COUNTRY POLICIES AND RANKING: SDG POLICY LANGUAGE Each country’s policies were obtained through ministerial websites, web searches and/or through ministerial contacts. No hard copies of policies were available. Only policies that were electronically accessible, whether through a government site, a partner site, or obtained via ministerial staff could be examined. Each country’s policies were analysed against the language of the following 13 health SDG targets: 1. By 2030, reduce the global maternal mortality ratio to less than 70 per 100,000 live births 2. By 2030, end preventable deaths of newborns and children under 5 years of age, with all countries aiming to reduce neonatal mortality to at least as low as 12 per 1,000 live births and under-5 mortality to at least as low as 25 per 1,000 live births 3. By 2030, end the epidemics of AIDS, tuberculosis, malaria and neglected tropical diseases and combat hepatitis, water-borne diseases and other communicable diseases 4. By 2030, reduce by one third premature mortality from non-communicable diseases through prevention and treatment

163

THE PUBLIC SPHERE â&#x20AC;&#x2020;| â&#x20AC;&#x2020;2017 ISSUE

and promote mental health and well-being 5. Strengthen the prevention and treatment of substance abuse, including narcotic drug abuse and harmful use of alcohol 6. By 2020, halve the number of global deaths and injuries from road traffic accidents 7. By 2030, ensure universal access to sexual and reproductive health-care services, including for family planning, information and education, and the integration of reproductive health into national strategies and programmes 8. Achieve universal health coverage, including financial risk protection, access to quality essential health-care services and access to safe, effective, quality and affordable essential medicines and vaccines for all 9. By 2030, substantially reduce the number of deaths and illnesses from hazardous chemicals and air, water and soil pollution and contamination 10. Strengthen the implementation of the World Health Organization Framework Convention on Tobacco Control in all countries, as appropriate 11. Support the research and development of vaccines and medicines for the communicable and non-communicable diseases that primarily affect developing countries, provide access to affordable essential medicines and vaccines, in accordance with the Doha Declaration on the TRIPS Agreement and Public Health, which affirms the right of developing countries to use to the full the provisions in the Agreement on Trade Related Aspects of Intellectual Property Rights regarding flexibilities

164

[Malawi, Rwanda, Uganda and Zambia - Sustainable Development Goalsâ&#x20AC;&#x2122; health targets]

to protect public health, and, in particular, provide access to medicines for all 12. Substantially increase health financing and the recruitment, development, training and retention of the health workforce in developing countries, especially in least developed countries and small island developing States 13. Strengthen the capacity of all countries, in particular developing countries, for early warning, risk reduction and management of national and global health risks (Sustainable Development Goals: 17 Goals to Transform our World, 2016)

METHODOLOGY FOR CALCULATING PREPAREDNESS Individual policy preparedness ranking This section looks at the code used to rank the countriesâ&#x20AC;&#x2122; policies based on preparedness and describes how the scores are calculated. It then discusses how results are fed into both collective policy and specific target preparedness rankings. For individual countries, there are five indicators and each indicator is ranked on a scale from 0-5. Table 1 Rankings Ranking code N/A 0 1 2 3 4

Not applicable to the country No/no evidence Vague association Evidence Clear evidence Similar wording

Very evident/explicit

165

THE PUBLIC SPHERE | 2017 ISSUE

Scores are calculated out of 325 based on the 5 indicators and 13 targets (5 points max per target x 13 targets x 5 indicators = 325 max). The maximum total drops by 25 points for each target that is not applicable to a country. For example, where the tobacco target is not applicable, the score is out of 300 instead of 325. Where both the tobacco and R&D targets are not applicable, the final score is calculated out of total of 275. This scoring system is used for its straightforwardness. For simplicity, all targets are weighted equally. The final results are divided by the total and multiplied by 100 (total/max*100 = overall preparedness per cent). The percentage equals the total level of preparedness for the country.

Collective policy preparedness ranking The results from individual countries feed into the collective policy preparedness ranking. In this matrix, each countries’ total score per indicator is calculated out of a maximum of 65 (5 points x 13 targets). For targets that are not applicable, the maximum number drops by 5. Formula to rank each countries’ performance by indicator: indicator total divided by country maximum multiplied by 100 (country’s indicator total/max*100=indicator preparedness per cent).

Specific target preparedness This allows the reader to understand at a cursory glance how a country is ranking in each target and where other countries rank for that individual target. This entails simply adding the totals from the 0-5 indicator ranking (no final percentages). For this matrix, the lowest score possible is 0 and the highest score possible is 25 (5 points x 5 indicators).

166

[Malawi, Rwanda, Uganda and Zambia - Sustainable Development Goals’ health targets]

RESULTS Context of results Donor policy in global health Many global health and development policies stem from international conferences. The most important of these are the SDGs and the Millenium Development Goals (MDGs), which will be analysed. The Millennium Development Goals (MDGs) were introduced in 2000 at the United Nations Millennium Summit and were intended to be achieved by 2015. The MDGs were the world’s eight “time-bound and quantified targets for addressing extreme poverty in its many dimensions-income poverty, hunger, disease, lack of adequate shelter, and exclusion-while promoting gender equality, education, and environmental sustainability” (UN Millennium Project, 2006). Goals 4, 5 and 6 directly addressed health issues such as reducing child mortality, improve maternal health, and combating HIV/AIDS, malaria and other diseases. As a result of this and other national and international initiatives, maternal and child mortality, tuberculosis, HIV/ AIDS and malaria health policies in Malawi, Rwanda, Uganda and Zambia are well established. The United Nations Sustainable Development Goals (SDGs) were recently developed as the successor to the MDGs. In 2012, the UN Conference on Sustainable Development (Rio+20) was a forum for governments to continue building on the success made in the last fifteen years. Of the 17 goals and 169 targets, goal 3 focuses on good health and well-being with a target date of 2030. In addition to the health concerns addressed in the MDGs, several more are now recognised in the SDGs, including: mental health, substance abuse, universal health care, quality care and drugs, environmental health, human resources, health risks and road safety. Some goals, such as research and development in pharma-

167

THE PUBLIC SPHERE | 2017 ISSUE

ceuticals, are less applicable to low-resource countries due to lack of capacity and funding.

Individual country analysis: Malawi Health system snapshot Malawi has a heavy burden of communicable diseases, especially HIV/AIDS with evidence of a growing number of non-communicable disease (WHO, 2015). Stunting from malnutrition affects nearly half of all children under five years old (Wilson, 2014). A quarter of women who want to use family planning have an unmet need due to barriers to access (lack of support from their husbands, length of travel to health centres, etc.) (Health Policy Project, 2016). Lack of human resources is one of Malawi’s main challenges, along with inadequate funding, equipment and infrastructure (Malawi Country Cooperation Strategy: At a glance, 2014).

Narrative analysis The Road Map for Accelerating the Reduction of Maternal and Neonatal Morbidity and Mortality is a strategic plan to tackle maternal mortality rates (MMR) (Ministry of Health (Malawi) 2012). Decreasing morbidity for children under five is discussed in the 2011-2016 Health Sector Strategic Plan (HSSP), but in the context of acute respiratory infections, diarrheal diseases, malaria, malnutrition, and vaccinations (Ministry of Health (Malawi), 2011). Target 3 in the HSSP on non-communicable diseases (NCDs) acknowledges the increasing burden of NCDs, including mental health, but there are no objectives listed in final policies or plans. No policies were found on substance abuse or road safety. Sexual reproductive health (SRH) integration has an up-todate 2015-2020 National Sexual and Reproductive Health and

168

Are targets explicitly expressed in policy language?

Are policies easily accesible for stakeholders?

Are crosssectoral and/ or intersectional components included?

Are policies integrated into work plans?

Results: 132.5/275 *100 = 48 % level of preparedness

Source: Author.

Are policies addressing any WHO health system building blocks?

Target 1 MMR

Target 2 Under 5

Target 3 CDs

Target 4 NCDS

Table 2 Policy preparedness indicator matrix: Malawi

Target 5 Alcohol

Target 6 Road

Target 7 SRH

Target 8 UHC

15.5

2.5

Target 9 Environ.

N/A

Target 10 Tobacco

N/A

Target 11 R&D

Target 12 HRH

Target 13 Risks

[Malawi, Rwanda, Uganda and Zambia - Sustainable Development Goalsâ&#x20AC;&#x2122; health targets]

169

THE PUBLIC SPHERE | 2017 ISSUE

Rights and HIV and AIDS Integration Strategy and endeavours to collaborate with researchers from various institutions of higher learning (National Aids Commission (Malawi), 2014). The universal health coverage (UHC) policy in the HSSP emphasises overall financial resources to the sector by strengthening health financing capabilities and working with the finance ministry to meet objectives. However, this overlooks financial risk protection for individuals. It also does not include any mention of access to medicine. There is evidence in the HSSP of environmental and health sector collaboration on air and water, but the 2004 National Environmental Policy does not have language on chemical and soil pollution (Ministry of Natural Resources and Environmental Affairs (Malawi), 2004). The WHO Framework Convention on Tobacco Control was not signed or ratified leading to no score. The HSSP has explicit policy language on the human resources for health target but does not offer concrete methods to increase health financing for human resources for health (HRH, p.38). As mentioned, the UHC policy included health financing language but it is unclear from the policy language if adequate resources will be allocated for HRH objectives. The policy does at least demonstrate plans for collaboration with the education ministry and regulatory bodies. There is strong policy language for strengthening the response to health risks in the HSSP, but Malawi lost points due to the HSSP not being easily accessible to stakeholders. Preparedness Score: 48 per cent.

Individual country analysis: Rwanda Health system snapshot The most common communicable diseases are HIV/AIDS, TB and diarrheal disease. Stunting of children under five remains high at 44 per cent (WHO, 2014). Health financing in Rwanda is characterised by the “Mutuelles de Santa” Community-Based

170

[Malawi, Rwanda, Uganda and Zambia - Sustainable Development Goals’ health targets]

Health Insurance scheme (WHO, 2014) which covers 90 per cent of the population. Rwanda’s high population density poses a burden for the health system in the capital, Kigali (Bahati, 2014). Regardless of the challenges, Rwanda’s health sector has been described as a “health care success story” (McNeil 1, 2013). See policy prepardness indicator matrix (Table 3) on following page.

Narrative analysis Rwanda’s 2012 Family Planning Policy covers MMR and neonatal deaths and is in the 2012-2018 third Health Sector Strategic Plan (HSSP III) (Ministry of Health (Rwanda), 2012). These documents contain some language on morbidity and mortality of children under five. According the HSSP III, the ministry’s Maternal and Child Health Unit will develop a new Child Survival Strategic Plan. The 2015 Health Sector Policy covers the well-funded communicable diseases (TB, AIDS, malaria) while excluding hepatitis (Ministry of Health (Rwanda), 2015). Rwanda has a 2015 Non-communicable Diseases Policy, but the National Health Policy stated that NCD prevention and control services are not widely affordable or available (Ministry of Health (Rwanda), 2015). There is also a National Mental Health Policy whose year of publication is unclear (Ministry of Health (Rwanda), n.d.). The health ministry’s 2011-2015 Adolescent Sexual Reproductive Health and Right Policy (ASHR) addresses narcotics and alcohol (possibly due to links made between risky sexual behavior and substance abuse among adolescents), but does not address older demographics and includes no language on addiction prevention (Ministry of Health (Rwanda), 2011). The 2012 Public Transport Policy and Strategy for Rwanda adequately covers the road safety target and embraces the cross sectoral collaboration by calling for a land transport safety council that includes a num-

171

THE PUBLIC SPHERE â&#x20AC;&#x2020;| â&#x20AC;&#x2020;2017 ISSUE

Are policies integrated into work plans?

Are crosssectoral and/ or intersectional components included?

Are policies easily accesible for stakeholders?

Are targets explicitly expressed in policy language?

Are policies addressing any WHO health system building blocks?

Target 1 MMR

Target 2 Under 5

Target 3 CDs

Target 4 NCDS

Target 5 Alcohol

Target 6 Road

Target 7 SRH

Target 8 UHC

Target 9 Environ.

N/A

Target 10 Tobacco

N/A

Target 11 R&D

Target 12 HRH

Target 13 Risks

Table 3 Policy preparedness indicator matrix: Rwanda

Results: 199/275 *100 = 72 % level of preparedness

Source: Author.

172

[Malawi, Rwanda, Uganda and Zambia - Sustainable Development Goals’ health targets]

ber of different ministries (Ministry of Infrastructure (Rwanda), 2012). The SRH access and integration target is a major policy piece for the ministry of health (MoH) with strategies in the Economic Development and Poverty Reduction Strategy (Ministry of Finance (Rwanda), 2008), the HSSP III and a specific policy for adolescents (ASHR). For adolescent policies, the MoH will collaborate with the ministry of education to cover adolescent SRH in school health policy and curriculum. There is evidence of collaborative efforts in the health financing arena — the Government of Rwanda will collaborate with the private sector to increase taxes on air tickets, foreign exchange transactions and other goods to supplement the national health sector budget (Ministry of Health (Rwanda), 2015). Over the last few years, health financing mechanisms have helped decrease incidences of financial risk, though financial risk protection is not specifically mentioned in the National Health Sector Policy. The 2016 National Pharmacy Policy promotes affordable and accessible medicines (Ministry of Health (Rwanda), 2016). The 2003 Environmental Policy endeavours to “formulate a national strategy for specific management of chemical products and biomedical and industrial waste,” but does not include explicit language on reducing deaths and illness (Ministry of Lands (Rwanda), 2003). The WHO Framework Convention on Tobacco Control was ratified in 2005 (WHO Report on the Global Tobacco Epidemic, 2015 Country profile Rwanda, 2015). In terms of the HRH target, there is language in the National Health Policy on collaborating with American university professors to train specialised doctors in Rwanda. There is substantive health risks policy in the 2012 Disaster Management Policy with explicit language on “mainstreaming” (incorporating policies) in education and cooperating with civil society organizations (Ministry of Disaster Management (Rwanda), 2012). Preparedness Score: 72 per cent.

173

THE PUBLIC SPHERE | 2017 ISSUE

Individual country analysis: Uganda Health system snapshot Uganda’s health system is highly decentralised and for political reasons, the number of districts has grown from 56 in 2005 to 112 in 2016. Financing, human resource, supply chain, and management issues hinder smaller health centres from achieving their mandates. Rampant corruption, a pharmaceutical black market, and acute human resource shortages are some of the health systems predominant challenges. The status of maternal health in the country has led to legal action against the Ugandan government for failing to provide adequate maternal health services consistently and effectively (The Guardian, 2015). It also has one of the highest rates of road accidents in the world, with recent claims of accidents causing more deaths than malaria and AIDS (China.org.Cn, 2016).

Narrative analysis Uganda’s 2007-2015 Roadmap for Accelerating the Reduction of Maternal and Neonatal Mortality and Morbidity in Uganda directly addresses both the maternal mortality and neonatal death rates (Ministry of Health (Uganda), 2007). The second National Health Policy (NHP II) states that the minimum health care package will feature child health, but does not elaborate further on the issue (Ministry of Health (Uganda), 2010). The national policy on NCDs has not been released but is in the finalizing stages. The Health Sector Development Plan 2015/2016-2019/2020 (HSDP) includes a section on NCD prevention and control but no language on mental health, resulting in a 2.5 ranking for policy language (Ministry of Health (Uganda), 2015). Ugandan media has noted the increase of narcotic drug abuse in the country but there is no reflection of prevention or treatment policy in the NHP II.

174

Are targets explicitly expressed in policy language?

Are policies easily accesible for stakeholders?

Are crosssectoral and/ or intersectional components included?

Are policies integrated into work plans?

Results: 157.5/300 *100 = 52.5 % level of preparedness

Source: Author.

Are policies addressing any WHO health system building blocks?

Target 1 MMR

Target 2 Under 5

Target 3 CDs

15.5

2.5

Target 4 NCDS

Table 4 Policy preparedness indicator matrix: Uganda

Target 5 Alcohol

Target 6 Road

Target 7 SRH

Target 8 UHC

Target 9 Environ.

Target 10 Tobacco

N/A

Target 11 R&D

Target 12 HRH

Target 13 Risks

[Malawi, Rwanda, Uganda and Zambia - Sustainable Development Goalsâ&#x20AC;&#x2122; health targets]

175

THE PUBLIC SPHERE | 2017 ISSUE

There is policy language in the HSDP to implement an Alcohol Control Policy and Drug Abuse Control Policy in the future. For the meantime, policy language on substance abuse is found in the 2004 Uganda National Adolescent Health Policy (Ministry of Health (Uganda), 2004). There are no road safety policy documents electronically available. For the SRH access and integration target, Uganda has the 2001 National Policy Guidelines and Service Standards for Reproductive Health (Ministry of Health (Uganda), 2001). UHC and access to quality care and medicines policy is well documented in the NHP II, HSSP, 2015 National Medicines Policy and 2015-2020 National Pharmaceutical Sector Strategic Plan (Ministry of Health (Uganda), 2015). There is a National Environmental Health Policy with the objective of creating an enabling environment for achieving and maintaining healthy living conditions (Ministry of Health (Uganda), 2005). There is no specific language on reducing deaths from various environmental hazards. The WHO Framework Convention on Tobacco Control was ratified in June 2007 (WHO Report on the Global Tobacco Epidemic, 2015 Country profile Uganda, 2015). The HSDP endeavours to implement the Tobacco Control Policy once the health ministry issues the 2015 Tobacco Control Act regulations. HRH is a vital issue for Uganda’s health sector and there is adequate language in the NHP II and HSDP. Uganda’s history of Ebola outbreaks (2012, 2011, 2007, 2000-2001) has helped build the country’s capacity to deal with health risks. Uganda’s 2010 National Policy for Disaster Preparedness and Management contains very specific language for handling pandemics and epidemics (Office of the Prime Minister (Uganda), 2010). Preparedness Score: 52.5 per cent.

176

[Malawi, Rwanda, Uganda and Zambia - Sustainable Development Goals’ health targets]

Individual country analysis: Zambia Health system snapshot Zambia has a large land mass and decentralised health system making equitable access of health services a challenge. To close this gap, religious organizations, most notably the Churches Health Association of Zambia (CHAZ), play an important role in delivering health services. CHAZ is the largest third sector health provider with 151 member institutions around the country (Chaz.org.zm, 2013). See policy prepardness indicator matrix (Table 5) on following page.

Narrative analysis One of the National Health Policy’s main objectives is to reduce newborn and maternal mortality (Ministry of Health (Zambia), 2012). In addition, the Ministry of Health published a 2013-2016 Maternal Neonatal and Child Health Road Map to address MMR, neonatal and child morbidity (Ministry of Health (Zambia), 2013). There are detailed plans in the NHP to tackle communicable diseases but there is no mention of hepatitis. For non-communicable diseases, there are policy measures in the NHP and National Health Sector Plan 2011-2015 (NHSP) to strengthen frameworks and coordination of services. Substance abuse policy is intertwined in the NHP with mental health policy and the 2011-2015 Adolescent Health Strategic Plan (Ministry of Health (Zambia), 2011). While it’s positive that substance abuse intersectionality is addressed with a specific plan for adolescents, the NHSP does not address substance abuse for older populations, resulting in a 2.5 score for work plan integration. No road safety policy was found for Zambia. There are many pol-

177

THE PUBLIC SPHERE â&#x20AC;&#x2020;| â&#x20AC;&#x2020;2017 ISSUE

Are policies integrated into work plans?

Are crosssectoral and/ or intersectional components included?

Are policies easily accesible for stakeholders?

Are targets explicitly expressed in policy language?

Are policies addressing any WHO health system building blocks?

Target 1 MMR

Target 2 Under 5

Target 3 CDs

Target 4 NCDS

15.5

2.5

Target 5 Alcohol

Target 6 Road

Target 7 SRH

Target 8 UHC

Target 9 Environ.

N/A

Target 10 Tobacco

N/A

Target 11 R&D

Target 12 HRH

Target 13 Risks

Table 5 Policy preparedness indicator matrix: Zambia

Results: 189.5/275 *100 = 68 % level of preparedness

Source: Author.

178

[Malawi, Rwanda, Uganda and Zambia - Sustainable Development Goalsâ&#x20AC;&#x2122; health targets]

icy documents addressing SRH access and integration, namely the Integrated Family Planning Scale-up Plan 2013-2020 (Ministry of Health (Zambia), 2013). Access to quality medicines is a listed objective of the NHP and the NHSP. The 2009 National Environmental Policy does not include explicit policy language on reducing deaths and illness (Ministry of Tourism (Zambia), 2009). The WHO Tobacco Framework on Tobacco Control was ratified May 2008 (WHO Report on the Global Tobacco Epidemic, 2015 Country profile Zambia, 2015). The target of supporting R&D of drugs is less applicable to low-income countries due to that burden typically falling on wealthier nations with the capacity to carry it out. Zambia has nonetheless managed to include pertinent language in the 2013 Guidelines for the Medicines and Therapeutics Committee (Ministry of Health (Zambia), 2013). The guideline calls for creating means to access medicines through the Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS). The NHP HRH policy is not explicit on financing or training. In response to the 2014 Ebola outbreak in west Africa, Zambia developed a 2015 Ebola Preparedness and Response Plan, ultimately strengthening its capacity for health risks (Ministry of Health (Zambia), 2015). Preparedness Score: 68 per cent.

Policy preparedness ranking comparison Comparative analysis Rwanda is the only country to have a perfect score on any policy preparedness indicators and ranks the highest overall, followed by Zambia, Uganda, and then Malawi. With the notable exception of Malawi, the countries examined are strongest in their accessibility of documents and knowledge management. The second highest ranked metric is explicit policy language. SRH integration is also highly ranked. The countries collective-

179

THE PUBLIC SPHERE | 2017 ISSUE

ly struggle most with cross-sectoral and/or intersectional policy language. All countries rank similarly on policies addressing WHO health system building blocks. Malawi’s bright spot is on integrating policies into work plans and using explicit policy language. However, Malawi is below par Table 6 Policy preparedness results and ranking matrix Malawi

Rwanda

Uganda

Zambia

Country w/ best results:

(25/55) *100 = 45 %

(23/55) *100 = 41 %

(28/60) *100 = 46 %

(23/55) *100 = 41.8 %

Uganda

(34.5/55) *100 = 62.7 %

(49/55) *100 = 89 %

(39.5/60) *100 = 65.8 %

(42/55) *100 = 76 %

Rwanda

Are policies easily accesible for stakeholders?

(5/55) *100 = 9%

(55/55) *100 = 100 %

(50/60) *100 = 83 %

(50/55) *100 = 90.9 %

Rwanda

Are crosssectoral and/or intersectional components included?

(23/55) *100 = 41.8 %

(29/55) *100 = 52.7 %

(9/60) *100 = 15 %

(32/55) *100 = 58 %

Zambia

Are policies integrated into work plans?

(45/55) *100 = 81.8 %

(43/55) *100 = 78 %

(31/60) *100 = 51.6 %

(42.5/55) *100 = 77 %

Malawi

48%

72%

52%

68%

Integration, Explicit targets, Addressing building blocks

Accessibility Explicit targets, Integration

Access., Explicit targets, Integration

Access., Integration, Explicit targets

Are policies addressing any WHO health system building blocks? Are targets explicitly expressed in policy language?

Average Countries strongest areas (top 3):

Source: Author.

180

[Malawi, Rwanda, Uganda and Zambia - Sustainable Development Goalsâ&#x20AC;&#x2122; health targets]

on accessibility of policies to stakeholders. Rwanda has a perfect score on policy access for stakeholders, followed by use of explicit policy language. Rwandaâ&#x20AC;&#x2122;s is lacking in cross-sectoral and intersectional components. Uganda scored best on policy accessibility, followed by explicit language, and struggles most with cross-sectoral and intersectional components of policies. Zambia has an almost perfect score on accessible policies, its best area, followed by integration of policies into work plans. Zambia is weakest in cross-sectoral and intersectional components in policy language but still outranks the other three countries in this area.

Preparedness to meet specific targets Comparative analysis Zambia scores top results on five SDG health targets: MMR, child health, communicable diseases, substance abuse and sexual reproductive health. The R&D target is not included in the final score but it is worth noting that Zambia is the only country that has policy language on finding ways to take advantage of the TRIPS agreement. Rwanda ranks second and is the only country with a road safety policy, and has the most robust policy on non-communicable diseases including mental health. Rwanda shares the highest score for universal health coverage. It rounds out with the best score for human resources for health policy. Uganda comes in third overall, sharing the top score with Rwanda in universal health coverage. Uganda ratified the Tobacco Framework on Tobacco Control and is the only country found to have established national tobacco objectives beyond legislation. Uganda also shares the top spot on global health risks policy. Malawi scores top on environmental health policy and ties with Uganda on highest ranking for global health risks policy. Collectively, all countries did similarly on sexual reproductive health integration, universal health coverage, financial risk protection

181

THE PUBLIC SPHERE â&#x20AC;&#x2020;| â&#x20AC;&#x2020;2017 ISSUE

and access to quality medicines, and human resources for health policy language. Table 7 Preparedness to meet specific targets Target number

Malawi results

Rwanda results

Uganda results

Zambia results

Top results

1. MMR

Zambia

2. Under 5

Zambia

3. CDs

Zambia

4. NCDS

15.5

Rwanda

5. Alcohol

15.5

Zambia

6. Road

Rwanda

7. SRH

Zambia

Rwanda, Uganda

9. Environ.

15.5

Malawi

10. Tobacco

Uganda

11. R&D

Zambia*

12. HRH

Rwanda

Malawi, Uganda

8. UHC

13. Risks

*Not included in final score. Source: Author.

POLICY RECOMMENDATIONS Malawi Stakeholders need easy access to the Health Sector Strategic Plan and, above all, the finalised version of the National Health Policy. It is essential that Malawi improve its knowledge man-

182

[Malawi, Rwanda, Uganda and Zambia - Sustainable Development Goalsâ&#x20AC;&#x2122; health targets]

agement capacities across government sectors. It is not for lack of ability as there are government units, albeit a limited number, that have managed to ensure stakeholdersâ&#x20AC;&#x2122; accessibility to policy documents. More specific objectives for non-communicable diseases should be adopted. There needs to be policy objectives on substance abuse and road safety. In terms of universal access to care, an emphasis on financial risk protection is necessary considering the limited disposable income of many Malawians. Given the dire state of supply chain logistics in Malawi, it is imperative that there is language to address access to medicines. Even if Malawi is less industrialised, there should be environmental health policies addressing soil and chemical pollution. This can be seen as a preventative measure in anticipation of emergent industries. As a low-income country, the R&D target is not expected to be the onus of Malawi but there can be policy language in place outlining how to identify or create the appropriate mechanisms to access medicines through TRIPS agreements. So long as tobacco remains Malawiâ&#x20AC;&#x2122;s top export, ratification of tobacco treaty is unlikely to happen. To make policy more easily accessible, Malawi must invest in information technologies and the capacity to make policies electronically available. This may come in the form of designated knowledge management personnel to ensure the most updated policies are circulated and posted. It will also benefit the ministry of health to upgrade the IT skills of those responsible for document management. Technical capacities such as health economics are needed to formulate a more specific and tailored policy on financial risk protection.

Rwanda Rwanda performed the best overall in the matrices. Rwanda scored highest in its human resources policy followed closely by its UHC and medicine access policy language, then SRH inte-

183

THE PUBLIC SPHERE | 2017 ISSUE

gration language. Additionally, Rwanda scored high for policies (NCDs and mental health, road safety policy) that are outside the already well-established MDGs 4, 5, and 6. Rwanda did best on the policy preparedness indicators, scoring highest on two out of the five indicators. It is the only country with more than one top score on the policy preparedness matrix. Even Rwanda’s weakest area (cross-sectoral and intersectionality) outperformed all countries in the category. However, Rwanda is weakest in environmental health policy and has yet to implement specific national government objectives in tobacco control. Communicable disease policy should include hepatitis. Prevention and control services for NCDs should be expanded. Existing services need to be made affordable and accessible to all. Given the amount of the population suffering from post-conflict mental illness, substance abuse policy cannot be partial to adolescents and should also address older demographics. UHC needs a more explicit focus on financial risk protection. Environmental policy needs more focus on reducing deaths and illness from chemical products. This needs to be in parallel with formulating a management strategy. Rwanda should endeavour to quickly translate any existing tobacco laws into action plans. NCD services need treatment and diagnostic technologies and increased human resource capacities. The remainder of the targets will require political will and specialised skillsets. Fortunately, Rwanda’s Ministry of Health has already made specialised training a priority.

Uganda Uganda scored highest on UHC and access to medicine. It then tied for MMR, SRH integration, HRH policy and global health risks. It also scored highly for NCDs and mental health policy. Uganda lacks an electronically accessible road safety policy and scored low on substance abuse and environmental health

184

[Malawi, Rwanda, Uganda and Zambia - Sustainable Development Goalsâ&#x20AC;&#x2122; health targets]

policy. Uganda had a second to highest score and three-way tie with SRH integration, HRH policy and global health risks. Given the high incidences of road injuries and mortalities in Uganda, road safety policy is imperative and should be the Ugandan governmentâ&#x20AC;&#x2122;s first priority. The National Health Policy needs more explicit language on addressing under-five child health. Drafting a separate policy similar to Rwandaâ&#x20AC;&#x2122;s can work as well. The national policy on NCDs that is currently being finalised will hopefully include language on mental health. There needs to be updated narcotics abuse language that addresses both older users and adolescents. Environmental health policy needs language on decreasing deaths from hazards. The forthcoming Tobacco Control Policy needs to be cross-sectoral by addressing youths. Note that Uganda is the only country where the tobacco target will be applicable due to the fact that it is the only country with specific national objectives in tobacco control. Most of these policies have a solid foundation to continue building on. It is a matter of political will and capacity to see policy development through. The road safety policy will be the most labour intensive of these given that the Ministry of Works and Transportation will need to make this a cross-sectoral effort. Enforcement of these policies will require additional financial and human resources.

Zambia Zambia ranks second in policy preparedness and scores top for cross-sectoral and intersectional components to policies. It ties at the top with MMR and preventing deaths of children under five. Their SRH integration policy follows closely. Zambia then ties again with their communicable disease and UHC policies. Zambia scores highly but does not have any language on road safety and scores low on environmental health policy.

185

THE PUBLIC SPHERE | 2017 ISSUE

Zambia’s Ebola plan is robust but it remains to be seen if it is flexible enough to translate to other health risks. Policies on communicable diseases should include specific language on hepatitis. NCDs can benefit from cross-sectoral collaboration in all areas, not just on oral health as it currently stands. The health ministry should likewise expand cooperative efforts, especially for NCD prevention. The National Health Sector Plan needs to address substance abuse for older populations. When considering access to medicines policy, stock outs must be considered. The National Environmental Policy needs to be strengthened to include explicit policy language on reducing illness and deaths. Tobacco use is mentioned in the National Health Policy which is a promising. Nevertheless, policy makers must ensure the tobacco industry does not dominate the anti-tobacco policy making process. The HRH policy should include focus on financing and training. To execute policies, Zambia requires similar capacities to the other countries. For most SDG targets, a platform in which to expand on existing policies is in place. Correspondingly, the government will need adequate financial and human resources for a brand new, cross-sectoral road safety policy.

CONCLUSION Malawi, Rwanda, Uganda and Zambia are all prepared to meet the SDG health targets. While it’s evident particular countries are better positioned to achieve the targets, all four countries are leading in some manner: Rwanda is the only country with a road safety policy, Zambia has policy language for R&D and accessing pharmaceutical commodities through the TRIPS process, and Uganda has moved forward with tobacco legislation. Malawi lags in some areas but still manages to have the strongest environmental health policy. Another good indicator of success is

186

[Malawi, Rwanda, Uganda and Zambia - Sustainable Development Goals’ health targets]

when countries have well established policy plans outside of the entrenched health MDGs (MMR and neonatal, child mortality, HIV/AIDS, TB and malaria). Beyond financial and human resource capacity, political will is by far the most integral aspect of seeing achievement of these goals come to fruition. At the central level, heads of state and ministers must have a vision all stakeholders can rally behind. Leaders must be willing to support the needs of those implementing policies and to prioritise them beyond talking points and campaign events. Policies need to be unique to the country’s setting to allow those on the ground to follow this vision. This research can be used as a cursory glance at policy gaps to help guide advocates when lobbying policy makers or allocating resources.

REFERENCES Adolescent Health Strategic Plan 2011 - 2015. (2011). 1st ed. [ebook] Lusaka, Zambia: Ministry of Health, p.27. Available at: http://www.mcdmch.gov.zm/sites/default/files/ downloads/ADH%20Strategic%20Plan%202011-15.pdf [Accessed 27 Aug. 2016]. Adolescent Sexual Reproductive Health and Right Policy. (2011). 1st ed. [ebook] Kigali, Rwanda: Ministry of Health, p.14. Available at: http://www.moh.gov.rw/fileadmin/ templates/policies/ASRH_and_Right_policy.pdf [Accessed 26 Aug. 2016]. Bahati, P. (2014). Rwanda: High Population Density a Challenge for Kigali Health System. AllAfrica. [online] Available at: http://allafrica.com/stories/201406270181.html [Accessed 26 Aug. 2016]. BBC, (2016). Zambia profile - full overview. [online] Available at: http://www.bbc.co.uk/ news/world-africa-14112920 [Accessed 27 Aug. 2016]. The Challenge | Malaria No More. [online] Malaria No More. Available at: https://www. malarianomore.org/pages/the-challenge [Accessed 26 Aug. 2016]. Chaz.org.zm. (2013). About Us | Welcome to Churches Health Association of Zambia Website. [online] Available at: http://www.chaz.org.zm/?q=about_us [Accessed 27 Aug. 2016]. China.org.Cn, (2016). Feature: Uganda battles notorious road accidents. [online] Available at: http://www.china.org.cn/world/Off_the_Wire/2016-08/25/content_39159860.htm [Accessed 27 Aug. 2016]. Cia.gov. (2016). The World Factbook — Central Intelligence Agency. [online] Available at: https://www.cia.gov/library/publications/the-world-factbook/rankorder/2004rank. html [Accessed 26 Aug. 2016].

187

THE PUBLIC SPHERE | 2017 ISSUE

Country Cooperation Strategy: at a glance Zambia. (2014). 1st ed. [ebook] World Health Organization. Available at: http://www.who.int/countryfocus/cooperation_strategy/ ccsbrief_zmb_en.pdf [Accessed 27 Aug. 2016]. countryeconomy.com. (2014). Rwanda - Population 2016. [online] Available at: http:// countryeconomy.com/demography/population/rwanda [Accessed 26 Aug. 2016]. Dieye, A. (2014). Socio-economic impact of the Ebola Virus Disease in Guinea, Liberia and Sierra Leone. 1st ed. [ebook] United Nations Development Programme, p.7. Available at: http://www.africa.undp.org/content/dam/rba/docs/Reports/UNDP%20Policy%20note%20EN_web.pdf [Accessed 26 Aug. 2016]. Disaster Management Policy. (2012). 1st ed. [ebook] Kigali, Rwanda: Ministry of Disaster Management and Refugee Affairs. Available at: http://midimar.gov.rw/uploads/ tx_download/Disaster_Management_Policy_01.pdf [Accessed 27 Aug. 2016]. Ebola Preparedness and Response Plan. (2015). 1st ed. [ebook] Lusaka, Zambia: Government of Zambia. Available at: http://www.moh.gov.zm/docs/reports/Final%20 Zambia%20Ebola%20Preparedness%20and%20Response%20Plan%202014.pdf [Accessed 27 Aug. 2016]. Economic Development and Poverty Reduction Strategy. (2008). 1st ed. Kigali, Rwanda: Ministry of Health, p.23. Environmental Policy. (2003). 1st ed. [ebook] Kigali, Rwanda: Ministry Of Natural Resources, p.6. Available at: http://www.minirena.gov.rw/fileadmin/Environment_Subsector/Laws__Policies_and_Programmes/Policies/POLITIQUE_ENVIRON-__Anglais.pdf [Accessed 26 Aug. 2016]. Family Planning Policy. (2012). 1st ed. [ebook] Kigali, Rwanda: Ministry of Health, p.6. Available at: http://www.moh.gov.rw/fileadmin/templates/Docs/Rwanda-Family-Planning-Policy.pdf [Accessed 26 Aug. 2016]. Family Planning Services: Integrated Family Planning Scale-up Plan 2013-2020. (2013). 1st ed. [ebook] Lusaka, Zambia: Ministry of Community Development, Mother and Child Health. Available at: http://www.healthpolicyproject.com/ns/docs/CIP_Zambia.pdf [Accessed 27 Aug. 2016]. Ganthu, J. (2016). Malawi: Harmful Cultural Practices Resurface and Threaten Malawi’s HIV Response. AllAfrica. [online] Available at: http://allafrica.com/stories/201601141275.html [Accessed 26 Aug. 2016]. Gender and Employment in Uganda. (2008). Ministry of Finance, Planning and Economic Development, (Briefing #9). Guidelines for the Medicines and Therapeutics Committee. (2013). 1st ed. [ebook] Lusaka, Zambia: Government of Zambia. Available at: http://www.moh.gov.zm/docs/ reports/guidelines_for_pharmacy_and_therapeutics_committee_2013.pdf [Accessed 27 Aug. 2016]. Healthpolicyproject.com. (2016). Malawi. [online] Available at: http://www.healthpolicyproject.com/index.cfm?id=country-Malawi [Accessed 26 Aug. 2016]. Health Sector Policy. (2015). 1st ed. [ebook] Kigali, Rwanda: Ministry of Health, p.5. Available at: http://www.moh.gov.rw/fileadmin/templates/policies/Health_Sector_Policy___19th_January_2015.pdf [Accessed 26 Aug. 2016]. Human Rights Watch. (2013). Zambia: Safety Gaps Threaten Copper Miners. [online] Available at: https://www.hrw.org/news/2013/02/20/zambia-safety-gaps-threaten-copper-miners [Accessed 27 Aug. 2016].

188

[Malawi, Rwanda, Uganda and Zambia - Sustainable Development Goalsâ&#x20AC;&#x2122; health targets]

Imbuwa, E. (2015). A Situation Report on Child Marriages in Zambia. [online] World Vision International. Available at: http://www.wvi.org/zambia/article/situation-report-child-marriages-zambia [Accessed 26 Aug. 2016]. June 2012-June 2018 third Health Sector Strategic Plan. (2012). 1st ed. [ebook] Kigali, Rwanda: Ministry of Health, p.30. Available at: http://www.moh.gov.rw/fileadmin/ templates/Docs/HSSP_III_FINAL_VERSION.pdf [Accessed 26 Aug. 2016]. Kambalame, E. (2013). Are youth friendly services friendly?. The Nation. [online] Available at: http://mwnation.com/are-youth-friendly-services-friendly/ [Accessed 26 Aug. 2016]. Malawi 2004 National Environmental Policy. (2004). Lilongwe, Malawi: Ministry of Natural Resources and Environmental Affairs. Malawi 2015-2020 National Sexual and Reproductive Health and Rights and HIV and AIDS Integration Strategy. (n.d.). Lilongwe, Malawi: Ministry of Health, p.20. Malawi Country Cooperation Strategy: at a glance. (2014). 1st ed. [ebook] World Health Organization. Available at: http://www.who.int/countryfocus/cooperation_strategy/ ccsbrief_mwi_en.pdf [Accessed 26 Aug. 2016]. Malawi Health Sector Strategic Plan 2011-2016. (2011). 1st ed. [ebook] Lilongwe, Malawi: Ministry of Health, p.65. Available at: http://www.gov.scot/resource/ doc/243072/0123901.docx [Accessed 26 Aug. 2016]. McNeil, D. (2013). Rwandaâ&#x20AC;&#x2122;s Health Care Success Story. The New York Times. [online] Available at: 1. http://www.nytimes.com/2013/02/05/science/rwandas-health-caresuccess-story.html?_r=0 [Accessed 26 Aug. 2016]. Myict.gov.rw. (2016). Policies. [online] Available at: http://www.myict.gov.rw/ict/policies-and-regulations/policies/ [Accessed 26 Aug. 2016]. National Environmental Health Policy 2005. (2005). 1st ed. [ebook] Kampala, Uganda: Ministry of Health. Available at: http://library.health.go.ug/publications/service-delivery-public-health/environment-and-sanitation/national-environmental-health [Accessed 27 Aug. 2016]. National Environmental Policy. (2009). 1st ed. [ebook] Lusaka, Zambia: Ministry of Tourism, Environment and Natural Resources, p.15. Available at: http://zm.chm-cbd.net/ implementation/legislation/policies-related-environment-and-biological/national-policy-environment-npe/npe_main_body_2009.doc [Accessed 27 Aug. 2016]. National Health Sector Plan 2011-2015. (2011). 1st ed. [ebook] Lusaka, Zambia: Ministry of Community Development, Mother and Child Health, p.54. Available at: http://www. mcdmch.gov.zm/content/national-health-strategic-plan-nhsp-2011-2015 [Accessed 27 Aug. 2016]. National Medicines Policy 2015. (2015). 1st ed. [ebook] Kampala, Uganda: Ministry of Health. Available at: http://health.go.ug/download/file/fid/589 [Accessed 27 Aug. 2016]. National Mental Health Policy in Rwanda. (n.d.). 1st ed. [ebook] Kigali, Rwanda: Ministry of Health. Available at: http://www.moh.gov.rw/fileadmin/templates/Docs/Posted-National-Mental-health-Policy-1.pdf [Accessed 26 Aug. 2016]. National Health Policy. (2012). 1st ed. [ebook] Lusaka, Zambia: Ministry of Health. Available at: http://www.mcdmch.gov.zm/sites/default/files/downloads/National%20 Health%20Policy%20-%20Final.pdf [Accessed 27 Aug. 2016]. National Pharmaceutical Sector Strategic Plan. (2015). 1st ed. [ebook] Kampala, Uganda: Ministry of Health. Available at: http://library.health.go.ug/publications/med-

189

THE PUBLIC SPHERE â&#x20AC;&#x2020;| â&#x20AC;&#x2020;2017 ISSUE

ical-products-technologies/pharmaceuticals-and-drugs/national-pharmaceutical-sector [Accessed 27 Aug. 2016]. National Pharmacy Policy. (2016). 1st ed. [ebook] Kigali, Rwanda: Ministry of Health, p.15. Available at: http://www.moh.gov.rw/fileadmin/templates/policies/Pharmacy-Policy_Rwanda-2016.pdf [Accessed 26 Aug. 2016]. National Policy for Disaster Preparedness and Management. (2010). 1st ed. [ebook] Kampala, Uganda: Directorate Of Relief, Disaster Preparedness And Refugees. Available at: http://www.ifrc.org/docs/IDRL/Disaster%20Policy%20for%20Uganda.pdf [Accessed 27 Aug. 2016]. National Policy Guidelines and Service Standards for Reproductive Health. (2001). 1st ed. [ebook] Kampala, Uganda: Ministry of Health, p.9. Available at: http://library.health. go.ug/publications/service-delivery-public-health/health-education/national-policy-guidelines-and-service [Accessed 27 Aug. 2016]. Non-communicable Diseases Policy. (2015). 1st ed. [ebook] Kigali, Rwanda: Ministry of Health, p.3. Available at: http://moh.gov.rw/fileadmin/templates/policies/NCDs_ Policy.2015.pdf [Accessed 26 Aug. 2016]. Public Transport Policy and Strategy for Rwanda. (2012). 1st ed. [ebook] Kigali, Rwanda: Ministry of Health, p.12. Available at: http://www.kigalicity.gov.rw/IMG/pdf/Final_ Draft_on_Public_Transport_Policy_and_Strategy_for_Rwanda_5_September_2012. pdf [Accessed 26 Aug. 2016]. Return, resettlement and re-integration strategic plan for Lango and Teso sub regions. (2005). 1st ed. [ebook] Kampala, Uganda: Government of Uganda. Available at: https://www.brookings.edu/wp-content/uploads/2016/07/Uganda_Lango-Teso_ Plan_2005.pdf [Accessed 27 Aug. 2016]. Road Map for Accelerating the reduction of Maternal and Neonatal Morbidity and Mortality in Malawi. (2012). 1st ed. [ebook] Lilongwe, Malawi: Ministry of Health, p.7. Available at: http://www.healthynewbornnetwork.org/hnn-content/uploads/Malawi-Roadmap-for-Reducing-MN-mortality-2012.pdf [Accessed 26 Aug. 2016]. Roadmap for Accelerating the Reduction of Maternal and Neonatal Mortality and Morbidity in Uganda. (2007). 1st ed. [ebook] Kampala, Uganda: Ministry of Health, p.21. Available at: http://library.health.go.ug/publications/service-delivery/sexual-and-reproductive-health/roadmap-accelerating-reduction-maternal [Accessed 27 Aug. 2016]. Roadmap for Accelerating Reduction of Maternal, Newborn and Child Mortality, 20132016. (2013). 1st ed. [ebook] Lusaka, Zambia: Ministry of Community Development, Mother and Child Health. Available at: http://www.mcdmch.gov.zm/sites/default/ files/downloads/MNCH_Road%20Map.pdf [Accessed 27 Aug. 2016]. Sachs, J. (2012). From Millennium Development Goals to Sustainable Development Goals. [online] Thelancet.com. Available at: http://thelancet.com/journals/lancet/article/ PIIS0140-6736(12)60685-0/fulltext [Accessed 26 Aug. 2016]. Scalzo, K. (2014). The Rwandan Refugee Crisis: Before the Genocide. [online] Nsarchive. gwu.edu. Available at: http://nsarchive.gwu.edu/NSAEBB/NSAEBB464/ [Accessed 26 Aug. 2016]. Second National Health Policy. (2010). 1st ed. [ebook] Kampala, Uganda: Ministry of Health, p.16. Available at: http://library.health.go.ug/publications/leadership-and-governance-governance/policy-documents/second-national-health-policy [Accessed 27 Aug. 2016].

190

[Malawi, Rwanda, Uganda and Zambia - Sustainable Development Goals’ health targets]

Sustainable Development Goals: 17 Goals to Transform our World. (2016). Goal 3: Ensure healthy lives and promote well-being for all at all ages. [online] Available at: 1. http:// www.un.org/sustainabledevelopment/health/ [Accessed 26 Aug. 2016]. The Guardian, (2015). Uganda’s supreme court rules maternal health suit must be heard. [online] Available at: https://www.theguardian.com/global-development/2015/ oct/30/uganda-maternal-health-supreme-court-ruling [Accessed 27 Aug. 2016]. Third Health Sector Strategic Plan July 2012 – June 2018. (2012). 1st ed. [ebook] Kigali, Rwanda: Ministry of Health. Available at: http://www.moh.gov.rw/fileadmin/templates/Docs/HSSP_III_FINAL_VERSION.pdf [Accessed 26 Aug. 2016]. Uganda National Adolescent Health Policy. (2004). 1st ed. [ebook] Kampala, Uganda: Ministry of Health, p.16. Available at: http://www.youth-policy.com/Policies/UGAyouthpol2004.pdf [Accessed 27 Aug. 2016]. Uganda Health Sector Development Plan 2015/2016-2019/2020. (2015). 1st ed. [ebook] Kampala, Uganda: Ministry of Health, p.53. Available at: http://library.health.go.ug/ publications/health-workforce-human-resource-management/performance-management/health-sector [Accessed 27 Aug. 2016]. UN News Centre, (2009). High birth rates hamper development in poorer countries, warns UN forum. [online] Available at: http://www.un.org/apps/news/story.asp?NewsID=30365#.V7m0W5MrLV0 [Accessed 26 Aug. 2016]. United Nations Development Programme, (2007). Famine in Malawi: Causes and Consequences. Human Development Reports. [online] Available at: http://hdr.undp.org/ en/content/famine-malawi-causes-and-consequences [Accessed 26 Aug. 2016]. United Nations Sustainable Development. (2015). Wake-up call’ data shows 63 million adolescents out of school – joint UN agency report - United Nations Sustainable Development. [online] Available at: http://www.un.org/sustainabledevelopment/ health/w [Accessed 26 Aug. 2016]. Unmillenniumproject.org. (2006). UN Millennium Project | About the MDGs. [online] Available at: http://www.unmillenniumproject.org/goals/ [Accessed 26 Aug. 2016]. WHO Report on the Global Tobacco Epidemic 2015 Country profile Malawi. (2015). 1st ed. [ebook] World Health Organization. Available at: http://www.who.int/tobacco/ surveillance/policy/country_profile/mwi.pdf [Accessed 26 Aug. 2016]. WHO Report on the Global Tobacco Epidemic, 2015 Country profile Rwanda. (2015). 1st ed. [ebook] World Health Organization. Available at: http://www.who.int/tobacco/ surveillance/policy/country_profile/rwa.pdf [Accessed 27 Aug. 2016]. WHO Report on the Global Tobacco Epidemic, 2015 Country profile Uganda. (2015). 1st ed. [ebook] World Health Organization. Available at: http://www.who.int/tobacco/ surveillance/policy/country_profile/uga.pdf [Accessed 27 Aug. 2016]. WHO Report on the Global Tobacco Epidemic, 2015 Country profile Zambia. (2015). 1st ed. [ebook] Lusaka, Zambia: World Health Organization. Available at: http://www.who. int/tobacco/surveillance/policy/country_profile/zmb.pdf [Accessed 27 Aug. 2016]. WHO Western Pacific Region. (2016). The WHO Health Systems Framework. [online] Available at: http://www.wpro.who.int/health_services/health_systems_framework/ en/ [Accessed 26 Aug. 2016]. Who.int. (2014). Rwanda Country Cooperation Strategy: at a glance. [online] Available at: http://www.who.int/countryfocus/cooperation_strategy/ccsbrief_rwa_en.pdf ?ua=1 [Accessed 26 Aug. 2016].

191

THE PUBLIC SPHERE â&#x20AC;&#x2020;| â&#x20AC;&#x2020;2017 ISSUE

Who.int. (2014). Rwanda Country Cooperation Strategy: at a glance. [online] Available at: http://www.who.int/countryfocus/cooperation_strategy/ccsbrief_rwa_en.pdf ?ua=1 [Accessed 26 Aug. 2016]. Who.int. (2015). Malawi: WHO statistical profile. [online] Available at: http://www.who. int/gho/countries/mwi.pdf ?ua=1 [Accessed 26 Aug. 2016]. Who.int. (2016). WHO | National health policies. [online] Available at: http://www.who. int/nationalpolicies/en/ [Accessed 26 Aug. 2016]. Wilson, R. (2014). New Anti-Stunting Drive To Begin In Malawi. World Food Programme. [online] Available at: https://www.wfp.org/stories/promising-approach-end-stunting-malawi [Accessed 26 Aug. 2016]. World Health Organization. (2016). Gender. [online] Available at: http://www.who.int/ gender-equity-rights/understanding/gender-definition/en/ [Accessed 26 Aug. 2016]. World Health Organization. (2016). Who we are, what we do. [online] Available at: http:// www.who.int/about/en/ [Accessed 26 Aug. 2016]. Worldbank.org. (2016). Uganda Overview. [online] Available at: http://www.worldbank. org/en/country/uganda/overview [Accessed 27 Aug. 2016].

192

Equality of opportunity and fiscal policies: Where are we? Where do we need to be? David Ortan

MSc International Political Economy (Research), Class of 2016 London School of Economics and Political Science

ÂŠThomas Hawk

Photo modified to black and white

ABSTRACT This paper evaluates to what extent equality of opportunity for income acquisition is achieved through tax-and-redistribution systems in the case of nine European countries, plus Japan, using the 2011 PIAAC survey. For the purposes of this paper, equality of opportunity for income acquisition is realised when the effect of peopleâ&#x20AC;&#x2122;s circumstances on their income is neutralised through the tax-and-redistribution system, while the effect of choices remains untouched. Circumstances are seen as being arbitrary, so that people should not be held responsible for them, while the opposite is true for choices. Results show that continental European countries perform better than United Kingdom (UK) and Japan in terms of the degree to which equality of opportunity is achieved. Moreover, within Europe, Nordic countries outperform Southern ones when considering different labour supply elasticities.

[Equality of opportunity and fiscal policies]

INTRODUCTION Since the publication of “Capital in the 21st Century” (Piketty, 2013), the problems of wealth and income inequality have been back on the (political) agenda. These issues are not new to social scientists, but the importance of the topic of inequality has reemerged for academics and policy makers alike. In the last thirty years, income inequality has increased, most worryingly in the US and UK, but also in the welfare states of continental Europe (Piketty, 2013). However, the debate is far from over. There are important questions that remain to be answered, namely: What are the effects of inequality for any society? To what extent is inequality unjust or just? To what extent is it possible for welfare states to promote equality without ignoring other goals? This paper explores the question of how much inequality, if any, would be acceptable in a given society, both theoretically and empirically, and to what extent developed states have achieved a desirable level of equality. In line with Cohen (1989) and Roemer (1998), it is argued that equality of opportunity (EOp) is an attractive version of equality because it corrects the effects of circumstances on people’s outcomes in life, while also leaving room for individual effort and choices. Thus, EOp is distinct from equality of outcomes because it also cares about the process by which an outcome, in this case a particular level of income, is achieved, not just the outcome itself. This paper tries to measure whether or not EOp for income is achieved, even if in an involuntary fashion, through tax and redistribution systems in the case of nine European states, plus Japan. Among the different approaches to EOp, this paper considers only EOp for income acquisition as it is the one that can be directly addressed by tax and redistribution systems. Hence, the main question this paper aims to answer is: to what extent do fiscal regimes achieve EOp for income acquisition? This question was first asked by Roemer et al. (2003), which

195

THE PUBLIC SPHERE | 2017 ISSUE

1. The data set was first used in Jerrim and Macmillan (2014) for exploring a different topic.

196

is the model that the present paper follows. EOp captures the idea that the effects of people’s circumstances in life should be compensated through a redistribution scheme, leaving inequalities due to differential effort untouched. The distinction of what constitutes a circumstance in life and what constitutes a choice is an issue that is far from settled (see Cohen, 1989; Dworkin, 1989a; Nozik, 2013; Rawls, 2009) For the purposes of this paper, I only consider people’s socio-economic background as a circumstance. Using a new data set, Survey of Adult Skills (PIAAC)1, this paper adds to the empirical literature on EOp by investigating whether the initial results of Roemer et al. (2003), which used data from the early 1990s, but also of later studies (Dardanoni et al.,2006; Checchi & Peragine, 2010; Björklund, Jäntti & Roemer, 2012) are confirmed. Moreover, it evaluates the situation of Japan from the point of view of EOp for income acquisition, which, to my knowledge, has received no attention in the literature. The case of Japan deserves particular attention because it is an OECD country that has experienced late but fast economic development. I show that continental European countries perform better than UK and Japan in terms of the degree to which equality of opportunity is realised. Furthermore, within continental Europe, Nordic countries outperform Southern ones in two scenarios. The first one considers parent’s level of education as the only circumstance leading to income inequality, while the second one adds parents’ immigrant background. The paper is structured in four sections. First, I set the stage for EOp and say why this might be a desirable policy. This argumentation ignores a great deal of detail, but it clarifies to the reader the choices this paper makes with respect to the version of EOp that it considers, namely EOp for income acquisition. Likewise, in the first section I present the theoretical framework first derived in Roemer (1998) and then revisited in Roemer et al. (2003) and other subsequent papers (section 2.2). Second, a brief literature review that shows the empirical results of other studies.

[Equality of opportunity and fiscal policies]

Then, the paper measures to what extent redistribution systems correct the effects of circumstances, while the result of effort untouched. A redistribution system is proposed, in line with EOp and then this system is compared with the observed policies of the states here under analysis. Finally, the paper discusses and evaluates the consequences of implementing EOp in terms of efficiency losses/gains.

THEORETICAL FRAMEWORK The theoretical framework of this paper closely follows Roemer et al. (2003). However, the paper goes beyond this model when defining and analysing the different theoretical issues that belong to the realm of political philosophy. Hence, the present framework has two parts: i) a philosophical one, in which I make the case for EOp and draw the differences between this approach and others, like equality of outcomes or equality of resources; and ii) an empirical (mathematical-economic) one, in which I present and discuss the tools that this paper aims to use in order to describe the EOp ideal.

1.1 Equality of opportunity, a philosophical defence In political philosophy, there have been contended several conceptions regarding equality in last decades. A very useful starting point for this explanation is the assumption that people should be equal in something (Cohen, 1989). The subsequent question is: under what dimension should people be equal? I am going to start by discussing equality of outcomes, as this is one of the most intuitive forms of equality, but also it is the one which, because it was refuted, generated other proposals regarding the appropriate dimension which should be equalised or equalisandum (â&#x20AC;&#x153;that which ought to be equalisedâ&#x20AC;? (Cohen, 1989, p.908)). Many claims have been raised against this conception of equality.

197

THE PUBLIC SPHERE | 2017 ISSUE

2. For a more detailed argument, see Cohen (1989) and Dworkin (1981a).

3. A similar critique can be made with respect to innate ability. People with more innate ability would require fewer resources than people with less innate ability in order to achieve a certain outcome. However, one cannot hold people responsible for their talents/ handicaps (Cohen, 1989).

198

Equality of outcomes entails making everyone equal with respect to a particular outcome regardless of their individual effort. One of the problems with equality of outcomes is that of “expensive tastes” (Keller, 2002). Equality of outcomes would entail giving more resources to people with expensive tastes than to those with less expensive tastes in order to satisfy their tastes and thus make them equally happy with those with those that can be satisfied with less expensive resources (goods, services etc.). Thus, people with expensive tastes would use a lot of resources for an arbitrary reason, namely that they have other tastes. In other words, equality of outcomes destroys the role of incentives in people’s lives, as it treats people in the same way irrespective of their actions.2 However, “expensive tastes” are not enough to defeat any compensation for unfavourable outcomes. If people with expensive tastes have these tastes through no fault of their own, then holding them responsible for these would seem unfair (Cohen, 1989). Even if he attacks equality of outcomes on the valid ground of expensive tastes, Dworkin (1981b) makes the unfortunate assumption that people “choose” their tastes and puts forward an alternative type of equality, namely equality of resources. The latter basically proposes to give equal endowments of resources to people, after which they are free to exchange them for goods according to their preferences/tastes. But Dworkin’s critique of equality of outcomes and his substitute are valid only in so far as people are responsible for their tastes. And there are multiple situations in life when people’s tastes are a result of upbringing and social environment, for example.3 Hence, in choosing an appropriate equalisandum, one should look at whether people are in a particular situation because of their choice or because of luck. A short digression is needed here in order to clarify what kind of luck egalitarians care about. An important distinction regarding luck was introduced by Dworkin (1981b) between “brute luck” and “option luck”. The former is the

[Equality of opportunity and fiscal policies]

type of luck over which the individual has no influence. Dworkin (1981b) uses the example of a person struck by lightning while on a trip, where he could not possibly have known about that (all weather forecasts predicted very good weather). The latter is a situation that has an uncertain outcome, but in which the individual chose to enter. To illustrate it, Dworkin (1981b) uses gambling, where a person chooses to be in a situation that has a certain probability of failure and the person is aware of that probability. Hence, the aim of egalitarians is to compensate for the effects of brute luck, which is seen as unfair and exploitative, on outcomes, but not for the effects of option luck. This is when EOp must be introduced. EOp is concerned not just with the outcomes that individuals achieve in some realm of life, but also with the process of achieving them. This theory aims to separate the factors that led to a certain outcome into choice-related and circumstance-related ones (Roemer, 1998). The latter category is the one for which we cannot hold someone responsible, and the former represents the part of an individual’s outcome for which we hold that person responsible. Hence, EOp could allow for equality of outcomes if we come to the conclusion that everything in one’s life is a matter of circumstance, but if that is not the case, it also permits some inequalities to exist, namely those that are a result of “choice/effort”. There are two major conceptions on EOp. The first one establishes that there should be no discrimination when it comes to the chance of achieving a certain outcome. This view entails a very formal kind of EOp and it is, from a philosophical standpoint, the least demanding one. For example, university admission should be open to all people, regardless of their gender, race, religion etc. The second conception goes a step further. It entails not just formal EOp between people, but it also considers social, economic and natural circumstances that might actually unjustly prevent someone from achieving a desired outcome. People from

199

THE PUBLIC SPHERE | 2017 ISSUE

disadvantaged social backgrounds, for example, might not have the opportunity of going to university even if higher education is formally accessible to them because they do not have the resources, skills, or they are discouraged to do so due to their social background (for instance, none of their parents went to university, their social environment does not appreciate the returns to higher education.). In this paper, I focus on the second approach to EOp. I consider social and economic background as being something that individuals have no control over, and hence they cannot be held responsible for. The inclusion of the social and economic background in “circumstances” rather than “choices” can be criticised on both egalitarian and non-egalitarian grounds (Cohen, 1989). From an egalitarian point of view, one might say that the social and economic status of a person is not so much of a circumstance, but it is also choice related (for example people could work less than others and thus, they could have lower incomes), so the compensation would be unfair. The non-egalitarian rebuttal would say, for example, that compensating for the effect of one’s disadvantaged social position represents a high burden for any society. For the purposes of this paper, I consider as “circumstances” the following dimensions: the level of education and the immigrant status of the surveyed individuals’ parents. I suggest most of the critiques to the inclusion of these two dimensions would be made from a non-egalitarian perspective. Critics could argue that the efficiency losses involved in an EOp policy are too large and hence we should not implement this policy. An example of such efficiency losses is a lower labour supply as a result of income redistribution. Finally, I need to mention what this paper is going to use as equalisandum. Even if it is not entirely accurate from a philosophical standpoint, my empirical approach will concern “opportunity for income”. Not even “opportunity for welfare” would be the right equalisandum, as it fails to take into account many

200

[Equality of opportunity and fiscal policies]

aspects that influence one’s outcomes in life. Cohen (1989) proposes the use of “access to advantage”, which is much broader than opportunity for welfare or income, although it includes both. However, he does not give a proper account of what “advantage” should contain. In line with Roemer et al. (2003), I do not take “opportunity for welfare” to be the desired equalisandum because that would entail interpersonal utility comparisons. Additionally, if “opportunity for welfare” cannot be considered as equalisandum, then certainly “access to advantage” cannot either as advantage is not just immeasurable and ambiguous, but it also requires us to compare utilities between people. So, even if very imperfect, for the purposes of this paper, “opportunity for income” strikes a balance between philosophy, economics and policy making. Not only is income measurable, clear and does not ask for inter-personal utility comparisons, but also most other advantages observable in today’s society depend on income.

1.2 A formalised approach to EOp In order to formalise the approach to EOp that this paper takes, there are five concepts to define: objectives, circumstances, type, effort and instrument (Roemer et al., 2003). The objective (denoted by ν) is defined as the desired outcome for which EOp needs to be achieved. Circumstances are these factors that influence a certain outcome which are not chosen by individuals and thus, they are not to be held responsible for. Effort is the set of factors that influence an outcome that are due to individual choice and for which people are to be held responsible. An instrument is the policy tool that can be used to achieve EOp as described above. It is denoted by φ є Φ, where Φ is the set of instruments available. All individuals that have the same circumstances are taken as a type, denoted as t є T , where T is the set of types and t a specific type within the set. Hence, νt(π, φ) will be “the value of the objective at the πth quantile (where π

201

THE PUBLIC SPHERE | 2017 ISSUE

є [0,1]) of the distribution of the objective in type t, at the policy φ” (Roemer et al., 2003, p. 542). The aim of this method is to determine the “value of the instrument which equalises the value of the objective across types at any degree of effort” (Roemer et al., 2003, p. 542). Hence, if a person is at the πth quantile of the effort distribution of her type, she expanded the same amount of effort as someone who is also at the πth quantile of the effort distribution of another type and should hence be equally rewarded. In my paper, the type considered is initially the level of education achieved by one’s parents, and the instrument is a tax and transfer policy that should seek to redistribute income, which is taken as the objective. The policy chosen should make the distributions of income for different types as equal as possible. According to Roemer et al. (2003, p. 543), the following function should be solved: 1

MaxφєΦ ∫Mintvt(π,φ)dπ 0

(1)

The formula intends “to find that policy φ that maximizes the minimum level of advantage (ν), across all types (t), of individuals who expend the πth degree of effort for their type” (Roemer, 1998, p. 27). This policy is Rawlsian with respect to circumstance-related outcomes, as it maximizes the circumstance-related outcome of those worse off and it is utilitarian with respect to effort-related outcomes because it leaves untouched outcomes dependent on the level of effort invested by each individual. The individual income after taxes and redistribution (henceforth, post-fisc income) is described by the following function: y=(1-a)x+c, where x is the pre-fisc income (income before taxes and redistribution), c is a universal grant and a is the tax rate. In this model, everyone is taxed at the same rate, which is an important simplifying assumption. If Gφt(y) is the cumulative dis-

202

[Equality of opportunity and fiscal policies]

tribution function of post-fisc income, y, corresponding to type t and policy φ, then by inverting π=Gφt(vt(π,φ) (distribution of effort) and plugging the result into equation 1, we obtain: 1

MaxφєΦ ∫Mint(Gtφ)-1(π)dπ

(2)

The formula above maximizes “the average post-fisc income of the most disadvantaged type” (Roemer et al., 2003, p. 546) using a given redistribution policy. Next, by ascribing a somehow well-known quasi-linear utility function (u(y,L) = y-αL1+1/α) and substituting into it the income function (y=(1-a)x+c), the optimal labour supply can be found after differentiation (Roemer et al., 2003, p.547): � η) L(w, a, c)=[(1-α)w/α]

(3)

Where α=α(1+1/η) and η is the wage elasticity of labour sup� ply. By assumption, the median worker works one unit of time. � ηw1+η Hence, the function for pre-fisc income x = wL=[(1-a)/α] can be obtained. Government revenues all come from taxation of pre-fisc income x = wL at flat rate a, and government expenditures are c+g per capita, with g non-transfer (education, health, infrastructure etc.) and c transfer payments per capita, so a balanced budget requires: ∞

� η w1+η f(w)dw = c+g a∫ [(1-α)/α]

(4)

203

THE PUBLIC SPHERE | 2017 ISSUE

On the left hand side of equation (4) there are the per capita tax revenues and on the right hand side, there are the per capita expenditures. We can now solve for c: � η B - g B= ∫w1+η dF c = a[(1-α)/α]

(5)

Assume type 1 is the most disadvantaged and let A=w1+ηdF1. Now, we can substitute both x and c into the post-fisc income equation y=(1-a)x+c. Hence, the average post-fisc income of the worst type is given by:

� η A+a[(1-α)/α] � η B-g (1-a) [(1-α)/α]

(6)

Maximizing the above mentioned equation, the tax rate corresponding to equality of opportunity is then (Roemer et al, 2003, p. 548):

aEOp=Max[1-ηB/((1+η)(B-A)), 0]

(7)

If there is a significantly large difference between B and A, then the tax rate corresponding to EOp would be the first term in the above parenthesis. If B/A≤1+η, then the optimal tax rate would be 0. In that case, the only tax that citizens pay should be the one for covering the per capita government expenditures, g. abench is defined as the tax rate that would allow the government to have a per-capita spending of g, without any transfer payments (c=0).

204

[Equality of opportunity and fiscal policies]

Next, the following index is defined (Roemer et al., 2003, p. 548):

ν = (V2 - V1) / (V3 - V1)

(8)

V1 is the average post-fisc income of the worst-type individual at the no-transfer policy, V2 is the average post-fisc income of the worst type at the observed policy and V3 is the same figure corresponding to the EOp policy. Hence, ν=0 if the observed policy is the no-transfer policy and ν=1 if the observed policy is the EOp transfer policy. To measure the deadweight loss, Roemer et al. (2003, p. 549) define the following index:

ε = ∫ x(w; φEOp)dF(w) ∫x(w; φ)dF(w) �

(9)

Where the numerator is the average income under the EOp policy and the denominator is the average income under the observed policy. If ε<1, then there is a deadweight loss involved under the EOp policy.

LITERATURE REVIEW Although the literature on EOp is quite extensive, this review is limited to the empirical applications of the theory and their findings. Reviewing all the theoretical conceptualizations and the different empirical approaches is beyond the scope of this paper. I intend here to give an overview of the empirical results in the literature in order to provide some information

205

THE PUBLIC SPHERE | 2017 ISSUE

4. Belgium, France, Germany (East and West), Denmark, Great Britain, Italy, the Netherlands, Norway, Spain, Sweden, and the USA.

206

about the level of EOp that different countries around the world have achieved. Roemer et al. (2003) compared the degree to which EOp for income was achieved in eleven countries4 through a tax and redistribution system for which period of time. When taking just the level of education achieved by parents as a circumstance, the author found that the Scandinavian countries, the Netherlands, West and East Germany taxed even more than EOp would have required. Belgium taxed just about enough to achieve EOp and other countries underperformed. Similar results on France were obtained by Lefranc, Pistolesi and Tranoy (2009). Dardanoni et al. (2006) used the method of quantile regression to test the extent to which EOp for pre-fiscal income was achieved in the UK and the US. Using survey data collected between 1957 and 1992, they found that neither of the two countries went so far as to correct for the influence of native ability or years of education on income. However, including these two factors under circumstances means taking a rather extreme/strong view on EOp. Nonetheless, even when considering just parents’ level of education as a circumstance (a factor that presumably more people would see as being something for which one cannot be held responsible), the UK underperformed and the results for the US were ambiguous after performing some robustness checks. Checchi and Peragine (2010) analysed the case of Italy and after accounting for a relatively limited set of circumstances (region, parents’ education and sex) found out that about 20 per cent of income inequality in the 90s is due to inequality of opportunity. Likewise, focusing on Italy, Peragine and Serlenga (2008) found that the social background (family) of an individual is very important in determining the educational outcomes, inequality of opportunity having a stronger effect in the South (the less developed part of Italy) than in the North (the more developed part).

[Equality of opportunity and fiscal policies]

Björklund, Jäntti and Roemer (2012) studied EOp for income in Sweden and found that the great majority of inequality there is due to effort, even after accounting for a very complex and detailed set of circumstances, decomposing the population into 1,152 types. Ferreira and Gignoux (2011) analysed a sample of Latin American countries. The study found that indeed there are vast inequalities between citizens that can be explained by uneven circumstances. Also regarding less developed countries, Cogneau and Mesplé-Somps (2008) studied inequality of opportunity for income for a sample of five African countries: Ivory Coast, Ghana, Guinea, Madagascar and Uganda, which are all located in the Sub-Saharan region. They concluded that inequality of opportunity is highly correlated with inequality of income and that the former British colonies from the sample (Ghana and Uganda) have “a much higher intergenerational educational and occupational mobility than the three former French colonies” (Cogneau & Mesplé-Somps, 2008, p. 1).

DATA AND METHODOLOGY This paper uses data from the PIAAC survey, which is an OECD wide survey that collected data on 24 countries between 1 August 2011 and 24 March 2012 (in the majority of cases). A representative sample of 166,000 people aged between 16 and 65 participated in the survey. Because of data incompleteness in both the PIAAC survey and the data set regarding taxation (see next paragraph), I only use the data for ten countries (Belgium, France, UK, Italy, Norway, Japan, Denmark, Ireland, Finland and Spain) and exclude individuals below 24 years old, as their income is usually nil probably because most of them are still living with their parents and/or not working. Moreover, as it is mentioned in the theoretical framework, combinations of the observed tax rate and grant (aobs,cobs) are

207

THE PUBLIC SPHERE | 2017 ISSUE

needed in order to compare these with the tax and grant values that EOp would require. As PIAAC does not contain data on post-fisc income, the pairs of (aobs,cobs) cannot be estimated by simply regressing post-fisc income on pre-fisc income. I therefore use the values for aobs derived by Van der Linde and Groot (2015), who use the data on pre-fiscal and post-fiscal Gini coefficients from the Standardized World Income Inequality Database (SWIID). The assumption behind this exercise is that both PIAAC and the SWIID contain a sufficiently large sample and therefore the values for aobs from Van der Linde and Groot (2015) can be safely combined with the PIAAC data set. The assumption regarding the redistribution system is that everyone is taxed at a flat rate and then the proceeds are distributed equally to everyone, in the form of a universal fixed grant. The value of the universal grant, cobs, is calculated by multiplying the tax rate with the value of the mean income and then subtracting g. I use “General government final consumption expenditure (per cent GDP)” from the World Bank database (World Bank, 2015) as a proxy for abench and then determine the value of g. The variables used from PIAAC are the pre-fisc income (monthly earnings including bonuses for wage and salary earners and self-employed) and the education level of parents (see table 1). Individuals are initially grouped in three types depending on their parents’ education (the circumstance). Looking at Table 1, for all countries standard deviations are quite large for each variable. However, the within-type large standard deviations might be explained by the fact that only parents’ education is taken as a circumstance. When extending the set of circumstances, the within type variation significantly reduces. This paper considers such an extension by taking as an additional circumstance the immigration status of the parents in addition to their level of after obtaining the results from the simple framework. The values indicated in the table are close to the national averages for gross earnings as indicated by Eurostat

208

[Equality of opportunity and fiscal policies]

(Eurostat, 2015). The sizes of the samples also differ. In the case of Italy, there are 2,143 individuals, while the sample for Denmark contains 4,410 observations. The size of the sample is a constraint on the number of circumstances that can be taken into account and consequently on the number of types that can be considered. Unlike Roemer et al. (2003), I cannot only consider the male subsample in the data set, as such a restriction would considerably reduce the sample. Such a restriction based on male/ female would have been desirable because the elasticity of labour supply is quite different between genders (Evers, De Mooij & Van Vuuren, 2008). The main reason for this is that women are more involved in household work than men (Borjas, 2013). The EOp pairs of tax and grant (aEOp, cEOp) are calculated as described in the theoretical framework. Afterwards, the indicator for the degree of closeness to the EOp ideal (ν) and the efficiency index (ε) are computed in order to be able to formally compare the degree to which EOp is achieved and to see whether there are welfare losses attached to the EOp policy. Basically, after taxing people at an aEOp flat rate and then giving everyone a fixed grant cEOp the continuous distribution functions (CDFs) of income of different types should get very close to one another.

RESULTS This section presents the results of the methodology presented in the theoretical framework. With respect to the values for η, I used in turn 0.06, 0.09 and 0.21. The first two were also used by Roemer at al.(2003). The first two are usually more appropriate for men, while the latter appears as representative for men and women taken together (Evers, De Mooij & Van Vuuren, 2008). As is discussed in the previous sections, the higher the value we use for η, the lower aEOp is. Table 2 shows that when η=0.06 and just parents’ level of education is taken as a circumstance, the Nordic countries from

209

THE PUBLIC SPHERE | 2017 ISSUE

Table 1 Summary statistics pre-fiscal income Country

Sample size

ED1

Pre-fisc income-per type ED2

BEL

2,630

3,282.5 (5,417.2)

3,325.7 (3,877.7)

3,661.7 (2,711.7)

3,394.8 (4,308.20)

4,233

1,921.9 (2,876.3)

2,286.2 (3,034.8)

2,935.2 (3,516.6)

2,319.3 (3,127)

3,204

2,213 (2,381.5)

2,423.3 (1,846.9)

2,994.1 (3,115.1)

2,433.7 (2,393.3)

NOR

3,268

369,145 (34,598)

37,817 (20,210)

41,529 (31,306)

38,779 (28,626)

2,143

2,077.3 (3,281.6)

2,488.4 (2,257.6)

3,281.3 (4,260.7)

2,256.2 (3,169)

JAP

3,102

288,143 (228,879)

342,278 (452,870)

370,757 (312,841)

336,852 (365,618)

4,410

31,810 (31,474)

3,4561 (44,569)

40,379 (63,178)

35,281 (47,210)

IRL

2,944

2,851.6 (4,636.3)

3,429.9 (4,800.2)

4,096.3 (8,288.2)

3,305.4 (5,738.2)

FIN

3,275

3,085.7 (2,032.3)

3,153.7 (1,665.2)

3,386.5 (1,845.1)

3,164.5 (1,861.4)

ESP

2,550

1,629 (1,147.7)

1,955.4 (1,395.5)

3,637.1 (19,696.3)

1,945.2 (7,299.5)

ED3

Mean pre-fisc income

Source: Author. Notes: The income refers to monthly earnings including bonuses for wage and salary earners and selfemployed -expressed in local currency-, and pre-fisc income for three groups that differ depending on the level of education achieved by their parents -expressed on a scale of 1-3, with “1” meaning that neither parent has upper secondary education, “2” meaning that at least one parent has attained upper secondary or post-secondary-non-tertiary education and “3” meaning that at least one parent has attained tertiary education- and pre-fisc income per group- St. dev. for each variable is in parenthesis. So variables that end with “1”, for example, give the level of income for those whose parents are in group 1 according to the scale explained above. Standard deviations are included in parentheses.

the sample overtax. Moreover, in the case of Norway, Finland and Belgium, EOp would indicate just a lump-sum tax for paying for the non-transfer expenditure of the government. Spain, Japan and the UK are the worst performers in our sample, signalling that there is quite a low level of socio-economic mobility. In the case of these three countries the dead-weight loss

210

[Equality of opportunity and fiscal policies]

Table 2 EOp policy, η=0.06 Country

Observed tax rate (aobs)

Observed grant (cobs)

EOp tax rate (aEOp)

EOp grant (cEOp)

Notransfer tax rate (abench)

EOp Closeness Indicator (v)

Efficiency Index (ε)

BEL

0.56

1089.73

-814.75

0.24

1.05

0.51

691.15

0.67

1066.13

0.21

0.65

0.98

0.52

683.88

0.38

330.54

0.24

1.02

NOR

0.57

13378.78

-8531.40

0.22

1.05

0.46

584.35

0.29

194.32

0.20

1.02

JAP

0.46

85897.18

0.61

137620.64

0.20

0.62

0.98

0.60

11466.35

0.42

5457.72

0.27

1.02

IRL

0.64

1530.40

0.59

1347.64

0.18

1.01

FIN

0.58

1066.45

-759.49

0.24

1.05

ESP

0.49

556.34

0.65

878.96

0.20

0.63

0.98

Source: Author. Notes: Parents’ level of education considered as the only circumstance (3 types, OT-overtax).

from implementing EOp policy would not be too large, as the values of ε indicate. When η=0.095 and again the only circumstance considered is parents’ level of education, then the only countries underperforming are UK and Spain, although they are very close to the EOp ideal. The same is true for Japan. As I said above, the fact that more countries are overtaxing when η=0.09 than when η=0.06 was expected. Thus, when η more than triples (i.e. η=0.21), all the countries in the sample appear as overtaxing, as indicated in Table 3. As considering just a single circumstance as influencing one’s income is a limited approach, I present below the results of a more complex situation. Tables 4 and 5 show the results of the same methodology that has been applied so far, with the difference being that parents’ immigration status is also taken as a cir-

5. The corresponding table not reported in the paper.

211

THE PUBLIC SPHERE | 2017 ISSUE

Table 3 EOp policy, η =0.21 Country

Observed tax rate (aobs)

Observed grant (cobs)

EOp tax rate (aEOp)

EOp grant (cEOp)

No-transfer tax rate (abench)

EOp Closeness Indicator (v)

Efficiency Index (ε)

BEL

0.56

1089.73

-814.75

0.24

1.19

0.51

691.15

-487.05

0.21

1.16

0.52

683.88

-584.10

0.24

1.17

NOR

0.57

13378.78

-8531.40

0.22

1.19

0.46

584.35

-451.24

0.20

1.14

JAP

0.46

85897.18

-67370.34

0.20

1.14

0.60

11466.35

-9525.89

0.27

1.21

IRL

0.64

1530.40

-594.97

0.18

1.24

FIN

0.58

1066.45

-759.49

0.24

1.20

ESP

0.49

556.34

-389.05

0.20

1.15

Source: Author. Notes: Parents’ level of education considered as the only circumstance (3 types, OT-overtax).

cumstance considered worthwhile for compensation under the EOp policy. Individuals are grouped into two categories: i) those that have at least one immigrant parent and ii) those with neither parent being an immigrant. Thus, if so far this paper has worked with three types (basic model), corresponding to the three levels of education achieved by parents, and now it considers six types (three education levels of parents x two immigrant statuses of parents, extended model). I excluded Japan and Finland from the sample of countries considered for the extended model because less than 6 per cent of the PIAAC survey respondents from these two countries had immigrant parents. In the extended model, all countries are quite far away from the EOp ideal, with values of ν between 0.4 and 0.74, when η=0.06 (see table 4). The findings for France and UK, as they stem from the PIAAC data set, cannot be reported in the ex-

212

[Equality of opportunity and fiscal policies]

Table 4 EOp policy, η =0.06 Country

Observed tax rate (aobs)

Observed grant (cobs)

EOp tax rate (aEOp)

EOp grant (cEOp)

No-transfer tax rate (abench)

EOp Closeness Indicator (v)

Efficiency Index (ε)

BEL

0.56

1089.73

0.80

1904.47

0.24

0.57

0.95

0.51

691.13

SVC

0.21

0.52

683.79

SVC

0.24

NOR

0.57

13378.78

0.75

20706.17

0.22

0.65

0.97

0.46

584.35

0.85

1457.12

0.20

0.40

0.93

0.60

11466.35

0.80

18839.12

0.27

0.61

0.96

IRL

0.64

1530.40

0.81

2074.67

0.18

0.74

0.96

ESP

0.49

556.34

0.83

1230.96

0.20

0.45

0.93

FIN

0.58

1066.45

-759.49

0.24

1.20

ESP

0.49

556.34

-389.05

0.20

1.15

Notes: Parents’ level of education and their immigrant status considered as circumstances (6 types). SVC-severe crossing between the CDF of the most disadvantaged type and the rest, the methodology cannot be applied. Table 5 EOp policy, η =0.21 Country

Observed tax rate (aobs)

Observed grant (cobs)

EOp tax rate (aEOp)

EOp grant (cEOp)

No-transfer tax rate (abench)

EOp Closeness Indicator (v)

Efficiency Index (ε)

BEL

0.56

1054.47

0.38

454.41

0.24

1.08

0.51

691.13

SVC

0.21

0.52

683.79

SVC

0.24

NOR

0.57

13378.78

0.25

992.32

0.22

1.12

0.46

584.35

0.53

738.45

0.20

0.79

0.97

0.60

11466.35

0.40

4549.72

0.27

1.09

IRL

0.64

1530.40

0.41

761.12

0.18

1.11

ESP

0.49

556.34

0.49

558.98

0.20

1.00

FIN

0.58

1066.45

-759.49

0.24

1.20

ESP

0.49

556.34

-389.05

0.20

1.15

Tables 4 and 5 Source: Author. Notes: Parents’ level of education and their immigrant status considered as circumstances (6 types).

213

THE PUBLIC SPHERE | 2017 ISSUE

tended model, as there is severe crossing between the CDF of the worst-off type and the rest of the CDFs. Looking at Table 5, Italy seems to underachieve with respect to EOp even when η=0.21 (ν=0.79), which is in line with Roemer at al. (2003) and Checchi and Peragine (2010), although both studies considered a different set of circumstances. Under this rather high-elasticity scenario, Spain seems to be the closest to EOp. If countries are clustered with respect to their performance in terms of EOp, then it can be observed that in the basic model, when η=0.06 (see table 2), a division line can be drawn between

0.2

0.4

0.6

0.8

FIGURE 1 PRE-FISCAL CDFS OF INCOME FOR 3 TYPES, DENMARK

20000

40000 earnmthall cinc1 cinc3

60000

80000

cinc2

Source: Author. Notes: Parents’ level of education taken as a circumstance. “cinc1”-cumulative distribution of pre-fisc incomes of those whose parents did not finish secondary school (type1), “cinc2” “cinc3”-the same variable for types 2 and 3, respectively, monthly earnings -“earnmthall”- on the x-axis.

214

[Equality of opportunity and fiscal policies]

0.2

0.4

0.6

0.8

FIGURE 2 PRE-FISCAL CDFS OF INCOME FOR 3 TYPES, ITALY

2000

earnmthall cinc1 cinc3

4000

6000

cinc2

Source: Author.

the more social democratic continental Europe, with the exception of Spain (ν=0.63), and the more liberal UK (ν=0.65) together with Japan (ν=0.62). When considering the extended model (6 types), there is a North-South division line within Europe, with Spain6 and especially Italy at the bottom of the ranking. That is seen from the systematically lower value of ν that these two countries from the “South” have compared to the countries from the “North” (Norway, Denmark, Ireland, and Belgium). Unfortunately, the relatively limited sample of respondents from each country did not allow to divide them into even more types, as in the case of Italy and other countries the regional factor (for example) could have played an important role (see Checchi & Peragine, 2010). Italy, Spain and the UK underperform in most of the settings described above. This situation can be explained by the higher pre-fisc inequality between types compared to the case of the

6. When η =0.21, Spain can with some approximation be seen as taxing at the level of EOp.

215

THE PUBLIC SPHERE | 2017 ISSUE

0.2

0.4

0.6

0.8

FIGURE 3. PRE-FISC VS. EOP POST-FISC CDFS, UK, 3-TYPES, Η=0.06

2000

4000 cpostinc3 cpostinc1 cinc2

6000

8000

cpostinc2 cinc3 cinc1

Source: Author. Notes: “cinc1”-CDF of pre-fisc incomes of those whose parents did not finish secondary school (type 1), “cpostinc1”- CDF of EOp post-fisc incomes of those whose parents did not finish secondary school (type 1), etc.

Scandinavian countries, for example. For a visual inspection of the above mentioned aspect, Figures 1 and 2 plot the pre-fiscal CDFs of Denmark and Italy for the 3-type case. To illustrate what the EOp redistribution system involves in a more intuitive manner, Figure 3 plots both the pre-fisc and the EOp post-fisc CDFs of incomes for the 3-type scenario in the case of UK. The effect is similar for the other countries in the sample. One can see that the distance between the distributions is

216

[Equality of opportunity and fiscal policies]

reduced, so that every type faces an almost identical distribution of income after EOp is implemented.

CONCLUSION AND POLICY RECOMMENDATIONS This paper aims to study to what extent EOp for income acquisition is achieved in ten countries: all of them situated in Europe, except for Japan. It uses the methodology of Roemer et al. (2003) in order to separate the effect of “circumstances” from the effect of “effort/choices” on individual incomes. Unlike Roemer et al. (2003), this paper also includes women in the sample considered. This is done mainly due to data restriction. I also consider a higher elasticity of labour supply than the above mentioned paper, namely η=0.21. This value is better for mixed samples (Evers, De Mooij & Van Vuuren, 2008). The circumstances included in this study are basic socio-economic conditions that could influence one’s outcomes in life, namely: parents’ level of education and their immigrant status. I first calculate the conditions that the redistribution system must comply with in the situation when only the first circumstance, parents’ level of education, is considered (basic model, three types) and then I include the second one, parents’ immigration status (extended model, six types) and repeat the procedure for six countries. I use a set of relatively strong assumptions that everyone is taxed at a flat rate and receives a universal grant from the governments. Because of this, I cannot say anything about the distribution of the tax burden between social groups. However, according to Roemer et al. (2003), the results do not change too much when progressive tax rates are used, while the methodology is far more complicated.

217

THE PUBLIC SPHERE â&#x20AC;&#x2020;| â&#x20AC;&#x2020;2017 ISSUE

The continental European countries in the sample overtax in the basic model, i.e. they tax more than EOp requires, even when considering very low estimates for the labour supply elasticity, except for Spain. Hence, the traditional division line between more social-democratic continental countries and more liberal ones is reinforced. In the extended model, the picture changes, in the sense that there is a North-South division within Europe, with the two Mediterranean countries in the sample (Spain and Italy) underperforming, even when using the higher estimate for labour supply elasticity. This paper presents some country results that differ from other studies or for whom the framework of Roemer at al. (2003) has never been used. In particular, France seems to overtax in all scenarios of the basic model, which is different from the conclusions of Roemer et al. (2003) and Lefranc, Pistolesi and Tranoy (2009). Japan, which has not been studied so far, appears as underachieving in the basic model. However, the number of circumstances considered in this paper is limited, so the reader should take these results with a grain of salt. From the countries analysed here, I suggest that Italy, Spain, UK and Japan could be taxing more for egalitarian reasons. However, the extent to which they have escaped the 2008 economic crisis might severely impact their ability to tax more for redistributive motives. Furthermore, it is fundamental that they redistribute the taxes so collected in a way that reduces inequality of opportunity. More specifically, they need to redistribute the tax revenues such that the incomes of the most disadvantaged types of citizens increase. This paper indirectly underlines the important role that the education of parents plays in creating opportunities for income acquisition for children. Hence, public investment in education is crucial for achieving EOp. Moreover, this paper also points out that, in order to achieve the EOp ideal defended in this paper,

218

[Equality of opportunity and fiscal policies]

Spain and Italy need to redistribute more money to those individuals whose parents have a migrant background. Nonetheless, it all fundamentally depends on whether policy makers adopt the egalitarian goals presented here.

REFERENCES Björklund, A., Jäntti, M., & Roemer, J. E. (2012). Equality of opportunity and the distribution of long-run income in Sweden. Social choice and welfare, 39(2-3), 675-696. Borjas, G. J. (2013). Labor economics (6th ed.). New York: McGraw-Hill. Checchi, D., & Peragine, V. (2010). Inequality of opportunity in Italy. The Journal of Economic Inequality, 8(4), 429-450. Cogneau, D., & Mesplé-Somps, S. (2008). Inequality of opportunity for income in five countries of Africa. Research on Economic Inequality, 16, 99-128. Cohen, G. A. (1989). On the currency of egalitarian justice. Ethics, 906-944. Dardanoni, V., Fields, G. S., Roemer, J. E., & Puerta, M. L. S. (2006). How demanding should equality of opportunity be, and how much have we achieved?. Mobility and inequality: Frontiers of research from sociology and economics, 59-82. Dworkin, R. (1981a). What is equality? Part 1: Equality of welfare. Philosophy & Public Affairs, 185-246. Dworkin, R. (1981b). What is equality? Part 2: Equality of resources. Philosophy & Public Affairs, 283-345. Evers, M., De Mooij, R., & Van Vuuren, D. (2008). The wage elasticity of labour supply: a synthesis of empirical estimates. De Economist, 156(1), 25-43. Eurostat. (2015). Average annual gross earnings by occupation. Retrieved May 21, 2015, from http://appsso.eurostat.ec.europa.eu/nui/show.do?dataset=earn_gr_isco&lang=en Ferreira, F. H., & Gignoux, J. (2011). The measurement of inequality of opportunity: Theory and an application to Latin America. Review of Income and Wealth, 57(4), 622-657. Jerrim, J., & Macmillan, L. (2014). Income inequality, intergenerational mobility and the Great Gatsby Curve: is education the key? (No. 14-18). Department of Quantitative Social Science-Institute of Education, University of London. Keller, S. (2002). Expensive tastes and distributive justice. Social Theory and Practice, 28(4), 529-552. Lefranc, A., Pistolesi, N., & Trannoy, A. (2009). Equality of opportunity and luck: Definitions and testable conditions, with an application to income in France. Journal of Public Economics, 93(11), 1189-1207. Nozick, R. (2013). Anarchy, state, and utopia. Basic Books. Peragine, V., & Serlenga, L. (2008). Higher education and equality of opportunity in Italy. Research on Economic inequality, 16, 67-97. Piketty, T. (2014). Capital in the 21st Century. Harvard University Press. Pinker, S. (2003). The blank slate: The modern denial of human nature. Penguin. Rawls, J. (2009). A theory of justice. Harvard University Press.

219

THE PUBLIC SPHERE â&#x20AC;&#x2020;| â&#x20AC;&#x2020;2017 ISSUE

Roemer, J. E. (1998). Equality of opportunity. Harvard University Press. Roemer, J. E., Aaberge, R., Colombino, U., Fritzell, J., Jenkins, S. P., Lefranc, A., & Zubiri, I. (2003). To what extent do fiscal regimes equalize opportunities for income acquisition among citizens?. Journal of Public Economics, 87(3), 539-565. Van der Linde, D., Groot, L. (2015). Inequality and Redistribution: Who Rules? (Working Paper, Utrecht University School of Economics). World Bank (2015). General government final consumption expenditure (% of GDP). Retrieved May 11, 2015, from http://data.worldbank.org/indicator/NE.CON.GOVT.ZS

Please see publicspherejournal.com for online acknowledgements.

220

The Spanish airport system Critical evaluation of the effectiveness of the Spanish government’s management of public resources Ane Elixabete Ripoll-Zarraga

PhD visiting researcher and PhD candidate in Entrepreneurship and Management (2017) Universitat Autònoma de Barcelona (UAB)

ABSTRACT The Spanish government has invested significantly in airport infrastructure, owned and managed through a public company (AENA). Overall, there is a low level of traffic, overlapping airport service areas, and significant cross-subsidisation. In this study, stochastic frontier analysis is used to account for airports’ inefficiencies caused by both environmental factors and managerial decisions. This is a requirement to critically analyse the effectiveness of governments in managing public resources and the impact of their decisions. The decision-making process is fully centralised and highly standardised in the Spanish airport industry. AENA has legal entitlement to decide public policies not only regarding price and quality of the services provided by the airports, but also depreciation accounting policies. These are usually decided for the network as a whole, potentially breaching the matching principal and accrual convention. Additionally, disaggregated financial statements are not published. Airports’ expansion periods are clearly followed by increases in airport charges to airlines. These matters question the transparency of the published financial statements as a true and fair view of the Spanish airport industry. The absence of an independent regulator calls for an assessment of the actual airports’ efficiency levels. Results demonstrate significant inefficiencies and public losses due to high fixed costs (pegged to high infrastructure investment). The results also show inflexibility at the airport level for competition, which does not allow the system to financially recover unless more decision power is granted to airports’ managers.

[The Spanish airport system]

INTRODUCTION Social concern regarding the optimum balance between airports’ ownership structure and management forms has increased considerably in recent years. Historically, airports used to be considered natural monopolies, fully owned by governments and treated as public utilities. In the Spanish case, the pressure of local governments, professional bodies and the general public has been focused on requesting the transfer of individual airports’ management to local authorities. Indeed, evidence from other OECD (Organisation for Economic Co-operation and Development) countries with similar-size airport networks and political structures suggests that improvements in efficiency stem from both decentralisation and private management (Nombela, G. in Abertis, 2009). Further, individual management of airports is used in most wealthy European countries such as France, Italy, Germany, and the UK, as well as in Canada and the United States. In some big cities with a high frequency of commuters and air travellers, airports may have a unique management form (public sector or private sector) in order to ensure competition in the airport market (London, Paris, Rome and Milan). In the Spanish case, an initial process to change the traditional owner-management model started on 25th March 2009, when the Spanish government announced the creation of a subsidiary (EGAESA)1 to manage Spanish airports. This decision seemed like a step towards privatisation, since private investors, as well as local councils, could share in airport equity. Talks over this issue of spun-out ‘singular airports’ have also been carried out by the Spanish Government since January 2010. The idea is to create subsidiaries in order to manage big airports (Madrid and Barcelona) separately from the rest of the network, which would remain under government control, while allowing for the participation of regional and local councils as well as commercial professional bodies at the Board of Directors (Cambra de Comerç de Barcelona, 2010).

1. Empresa de Gestion de Aeropuertos del Estado

223

THE PUBLIC SPHERE | 2017 ISSUE

2. Airport charges are the fees paid by an airline to an airport in exchange for using the airport’s infrastructure. These are usually movementbased (runway; parking; terminal; noise, etc.) and passengers-based (services; security; infrastructure, etc.). Cargo charges are not enclosed since the service is usually outsourced. 3. A catchment area is understood as the area of influence of an airport to attract visitors and customers, depending on the population nearby and the surface transport possibilities (European Court of Auditors, 2014).

224

All managerial decisions are made at governmental level through AENA (government-owned and -managed) including airport charges2 and accounting policies. In order to understand the effects of the centralised decision-making process, it is necessary to assess its impact on airports’ operational activity (namely, traffic). The Spanish airport system is cross-subsidising, transferring financial resources from profitable to non-profitable airports. Non-profitable airports do not have enough traffic to cover their fixed costs. The question thus comes down to understanding if the smallest airports are actually required for connectivity purposes, or if they should be closed down. A similar airport system structure can be found in Norway, where a number of airports fall within the same catchment areas.3 In this instance, small airports remain open for ensuring connectivity in remote areas since there is not an alternative travel model: these are known as public services obligation routes (PSOs). The government establishes a minimum number of transport services, subsiding part of the cost incurred by airlines. One consequence of centralised management is that Spanish airports do not compete, even when located closed to each other. Competition starts with rivalry in the industry (Porter, 1979): Markets will become more competitive when they have lower barriers to entry. In the Spanish case, this translates into providing flexibility in the negotiation process with airlines regarding prices and quality of services. Spanish airports do not compete among themselves since airport managers do not have decision-making power over key management levers such as passenger choice, service provision, and airport fees. Spanish airports located in the same catchment area could compete if the necessary mechanisms existed (CNMC The National Board for Markets and Competition, 2014). The centralisation of the decision making-process has led to an excess of small airports unused (or underutilised) for aeronautical purposes, becoming a strain on public financial resources. These are the result of over-investment based on erroneous and outdat-

[The Spanish airport system]

ed forecasts (European Commission, European Court of Auditors, 2014). Excess of capacity also exists for medium and large airports: The vast majority of Spanish airports currently have idle infrastructure, making their system technically inefficient overall, as well as non-profitable. Furthermore, these suboptimal investment decisions, along with other unnecessary infrastructures such as the high-speed train, have represented an overspending of public resources traditionally funded with loan resources resulting in a significant public deficit4 (European Commission Press release, 2016; Word Finance, 2016). These matters call for an assessment of the efficiency of Spanish airports in using their current infrastructure capacity to develop traffic. Further, it becomes necessary to analyse whether non-profitable airports (for example, due to unnecessary expansion and the consequent low-level of traffic to meet the actual capacity) could, in the long-term, become profitable with a better usage of the infrastructure, by becoming technically efficient. This would determine whether airports’ excess capacity issues are due to an ineffective centralised management process or caused by the airports’ environment, such as geographical location. This excess capacity is termed simply “inefficiency” and is distinct from not making a profit; an airport may operate at full efficient capacity but not charge enough to break even. In this paper, Stochastic Frontier Analysis (SFA) is used in order to calculate airports’ inefficiencies, with the aim of distinguishing between controllable factors (inputs and outputs) and externality-induced factors. Airports are typically multi-output firms and consequently, an appropriate framework must be considered to estimate the efficiencies — hence the use of stochastic distance functions (Coelli & Perelman, 1999; Kumbhakar & Lovell, 2000). There are several stochastic frontier models for panel data, though few studies have applied them to empirical research (Kumbhakar et al. 2014). Previous studies have used other techniques such as data envelopment analysis (DEA), but the many shortcomings

4. The main airport activity (traffic) should generate enough revenues to cover not only the airports’ operating costs, but essentially their fixed costs, such as employees and depreciation of the assets, as well as the financial interest expense incurred due to investments. The commercial diversification has been evidenced in privatisation processes since commercial revenues become an important source of financing.

225

THE PUBLIC SPHERE â&#x20AC;&#x2020;| â&#x20AC;&#x2020;2017 ISSUE

identified encourage the use of SFA. The model used, proposed by Battese and Coelli (1995), allows for the inclusion of environmental variables. The environmental factors in SFA refers to externalities (positive and negative) beyond managerial control that may affect technical efficiency. The question to be addressed is, therefore, to what degree is airport inefficiency caused by management decisions, relative to external factors? In other words, this study asks whether these airportsâ&#x20AC;&#x2122; technical inefficiencies could be effectively reduced, or whether they arise from uncontrollable causes. The next section presents a review of the literature review on the Spanish airport efficiency, followed by a methodology section, description of the data, and formulation of the model. The final two sections present a discussion of the results and conclusions.

LITERATURE REVIEW In industries based on network services, such as air transport, the input distance function approach is more reliable compared to cost function estimation for describing performance (Coelli et al. 2005). The former allows the estimation of airport inefficiencies in the absence of input prices. It is also essential to account for environmental factors, such as region-specific demand conditions (Hattori, 2002). Although these are not under the control of airport managers, the obligation to supply services, given that airports are publicly-owned, and respond to the same issue of connectivity may be affecting the airport operational activity and maintenance. The environmental variables are initially assumed to be beyond management control, since they are not identifiable (unobserved heterogeneity). Therefore, these variables are unchangeable, at least in the short-term (within one year). Several translog models have been tested within stochastic frontier analysis for panel data for both time varying and invariant inefficiencies assuming different distribution of the inefficiency term. This

226

[The Spanish airport system]

is done in order to use a model that fairly represents the actual situation of the Spanish airport-system and draw sound conclusions regarding the impact of AENAâ&#x20AC;&#x2122;s management specifically. Previous studies in regards to the Spanish airport system differ slightly in their methodologies, main objectives and results. The literature shows a significant number of studies using non-parametric analysis such as data envelopment analysis (see for example Murillo-Melchor, 1999; Salazar de la Cruz, 1999; Martin & Roman, 2001; Tapiador et al., 2008; Lozano & Gutierrez, 2011; Lozano, 2013; Coto-Millan et al. 2014 and 2016). Few studies use parametric methodologies such as cost functions (Martin-Cejas, 2002) or stochastic frontier analysis (Coto-MillĂĄn et al. 2007; Martin et al. 2009 and 2011; Tovar & Martin-Cejas, 2009 and 2010). Parametric methods require the determination of a specific function prior to the analysis based on the relation between the variables. Non-parametric techniques do not require the definition a production function. Table 1 gives details of some relevant studies and their findings. Overall, the results show that airports with more traffic in terms of number of passengers are more efficient. Nevertheless, few studies look for reasonable explanations of inefficiencies (Coto-MillĂĄn et al. 2014; 2016). In fact, second-stage regressions explaining inefficiencies are not advised since they could mislead the results in data envelopment analysis (Silmar & Wilson 1999; 2007), as well as in stochastic frontier analysis (Battese & Coelli, 1995; Wang & Schmidt, 2002). One of the main issues is that regressors used to explain the inefficiencies are not identically distributed (Battese & Coelli, 1995) or correlated with the efficiency units (Simar & Wilson, 2007). The literature clearly concludes that it is necessary to look for further explanations of inefficiencies beyond the level of traffic alone. The fact that accountability and commercial policies are decided at the governmental level (AENA) for the whole network, rather than by individual airports managers, requires a critical assessment. A new perspective must be adopted.

227

THE PUBLIC SPHERE | 2017 ISSUE

Table 1 Summary of Spanish airports’ studies Author(s)

Outcome

Murillo-Melchor (1999)

Technical Efficiency 33 Spanish airports (1992-1994)

Salazar de la Cruz (1999)

Technical Efficiency 16 Spanish airports (1993-1995)

Martin & Roman (2001)

Technical Efficiency 37 Spanish airports (1997)

Data Envelopment Analysis (DEA)

Martin-Cejas (2002)

Technical Efficiency 40 Spanish airports (1996-1997)

Deterministic Cost Frontier (DCF)

Martin & Roman (2006)

Technical Efficiency 34 Spanish airports (1997)

Different variations based on Data Envelopment Analysis (DEA)

Coto-Millan et al. (2007)

Tapiador et al. (2008)

Martin et al. (2009)

Tovar & MartinCejas (2009)

228

Economic Efficiency 33 Spanish airports (1992-1994) Technical Efficiency 29 Spanish airports (2006-2007) Economic Efficiency 37 Spanish airports (1991-1997) Technical Efficiency 26 Spanish airports (1993-1999)

Methodology Data Envelopment Analysis (DEA) Total Factor Productivity (Malmquist Index)

Findings

Data Envelopment Analysis (DEA)

Airports with more passengers are more efficient

Airports with larger size are more efficient. Airports’ geographical location affects efficiency Airports with 1 to 3 million passengers show higher average of efficiency Airports with more passengers are more efficient. Airports’ geographical location affects efficiency

Cost Stochastic Frontier Analysis (SFA)

Airports with more passengers are more efficient

Data Envelopment Analysis (DEA)

Larger and small airports are more geographically efficient

Cost Stochastic Frontier Analysis (SFA) Bayesian Inference Stochastic Frontier Analysis (SFA) Distance Function

Larger airports are more efficient Airports outsourcing some services are more efficient

[The Spanish airport system]

Summary of Spanish airports’ studies (cont’d) Author(s)

Outcome

Tovar & MartinCejas (2010)

Technical Efficiency 26 Spanish airports (1993-1999)

Lozano & Gutierrez (2011)

Technical Efficiency 39 Spanish airports (2006-2007)

Passengers and Target-setting Cargo are directly DEA Slack-Based related with Measure (SBM) efficiency

Martin et al. (2011)

Economic Efficiency 36 Spanish airports (1991-1997)

Cost Stochastic Frontier Analysis (SFA)

Airports within the same catchment area are costinefficient unless congested

Lozano et al. (2013)

Technical Efficiency 39 Spanish airports (2008)

Network DEA

Network DEA shows higher discriminatory power to detect inefficiencies

Coto-Millan et al. (2014)

Technical Efficiency 35 Spanish airports (2009-2011)

Coto-Millan et al. (2016)

Technical Efficiency 35 Spanish airports (2009-2011)

Source: Author.

Methodology Stochastic Frontier Analysis (SFA) Distance Function. Total Factor Productivity (Malmquist Index)

Data Envelopment Analysis (DEA) Total Factor Productivity (Malmquist Index) Regression (Airport’s size; Low Cost Carriers (LCC)) Data Envelopment Analysis (DEA) Tobit Regression (Airports’ size; Cargo; LCCs)

Findings Hub airports are on average more efficient. Northern airports are more efficient

Larger airports are more technically and scale efficient LCC increases scale efficiency

Airports with more cargo are more technically and scale efficient.

229

THE PUBLIC SPHERE â&#x20AC;&#x2020;| â&#x20AC;&#x2020;2017 ISSUE

METHODOLOGY Parametric models define a specific production function compared to non-parametric models that do not require assuming a specific form (i.e. DEA). SFA overcomes some of the issues found when applying DEA. These include sensitivity to the number of inputs and outputs used relative to the number of observations (all the variables are considered to contribute to efficiency levels); the presence of outliers affecting the efficiency levels of the rest of the airports (i.e. Barcelona and Madrid); or the lack of dynamism and justification behind the efficiency scores (benchmarking against best practice). Two airports could be technically efficient, but this efficiency could be reached in different ways. Some airports are more cargo-specialised (for example Vitoria or Zaragoza) compared to others, which deal mostly with passengers; others focus on both services. In the same way, some of the inputs may have differential impacts on different airports depending on geography (e.g. population of the city or region, demographic characteristics, season, climate, etc.). SFA reflects the individual impact of the inputs (as well as outputs) that explain overall technical efficiency and those that, although incorporated, are not significant. On this basis, SFA becomes a more managerial tool compared to DEA. The calculation of SFA efficiencies requires the formulation of a parametric production function. The production function reflects the association between the variables across the observations: in this case inputs (resources) used to generate outputs (traffic). The production function is determined by the existing technology (infrastructure): the more resources an airport has (number and extension of runways; number of terminals; boarding gates; aprons; employees; desk counters; etc.) the higher the level of traffic that can be produced (passengers; air traffic movements and cargo). Different SFA models correspond to the assumptions made regarding the distribution of the inefficiency

230

[The Spanish airport system]

term as well as time-varying or invariant inefficiencies. In this study, Battese and Coelli (1992) has been used in order to account for time-varying inefficiencies (panel data) thus allowing for the introduction of environmental factors. Battese and Coelli (1992 and 1995) assume the expected value of the inefficiency term to be different from zero.5 The functional form of the stochastic production function may have different specifications, such as Cobb-Douglas (one output and several inputs) or its generalisation, the transcendental logarithmic (“translog”). Cobb-Douglas is a specific case of the translog function imposing additivity and homogeneity (i.e. constant elasticity of substitution). The translog function has a flexible functional form allowing for a multi-input-output case.6 That is, it does not impose restrictive assumptions regarding substitutability between resources and production levels. Consequently, the use of the translog production function is based on its properties of flexibility and homogeneity, allowing for partial elasticities of inputs-substitution to vary.7 The translog form of the stochastic distance function lnDit(X,Y) assuming m outputs and k inputs follows:

ln( X ) = β0 + ∑j=1 βj ln(x*jit) + 1/2 ∑j=1 ∑j' βjj' ln(x*jit) ln(x*j'it) ϰit

+ ∑l=1 αl ln(ylit) + 1/2 ∑l=1 ∑l'=1 αll' ln(ylit) ln(yl'it) k

+∑j=1 ∑l=1 βjαl ln(x*jit)ln(ylit)+(vit-uit)

(1)

This model attempts to explain the amount of a specific resource used (xj) by an airport i in a specific moment of time t based on the consumption of the rest of the inputs (x’j), which impact is considered individually (βj)8 as well as their interaction (βjj’). For example, the airports’ operating costs not only depend

5. Kumbhakar (1990) assumes half-normal distribution with zero mean and constant variance ui~N+(0,σ2) where uit =ui. [1+exp(bt+ ct2)]-1. 6. The Cobb-Douglas function follows yit=β0Πkj=1 xjitβj and the Translog yit=exp[β0+∑kj=1βjln(xjit) + 1/2∑kj=1∑kj'βjj’ln (xjit) ln(xj’it)]. 7. Note that the Cobb-Douglas function requires a constant elasticity of substitution equal to 1 being restrictive regarding elasticity of substitution and scale properties. This can be interpreted as the expansion path being a straight line from the origin (resources or goods that are not perfect complementaries, neither perfect substitutes). For more than one product or more than two factors of production, constancy of elasticities of substitution and transformation is highly restrictive. The translog function collects the iteration effects of different combinations between inputs, outputs and input-outputs. 8. The individual coefficients represent the inputs (βj) and outputs elasticities (αl).

231

THE PUBLIC SPHERE | 2017 ISSUE

on the number of employees, but also on how big the airport is in terms of infrastructure. This model also assumes that the operating costs depend on the different outputs produced (yl) and their iteration (αll’). The more traffic (passengers, air traffic movements and cargo) the higher the operation costs. Additionally, although there is an expected relation between the three outputs (more passengers should imply more air traffic movements andcargo), it does not always work in this way; some airports are cargo-oriented (few or zero passengers), compared to others being passenger-oriented, and this specialisation will influence the rest of the inputs (βjαl). Others may have a significant number of movements, but few passengers. In this sense, the number of employees not only depends on how big an airport is in terms of infrastructure (terminal buildings, checking desks, etc.), but also on its traffic (level of passengers and number of aircraft landing) and also their relation: more infrastructure, but less traffic, may require a lower number of employees similar to smaller airports. Following Lovell et al. (1994), an input has been chosen arbitrarily for normalisation purposes: x*jit = (xjit / xϰit). The translog form permits the imposition of homogeneity (Lovell et al. 1994): m

∑ l=1 αl = 1, ∑ l=1 αll' = 0, ∑ l=1 βjαl = 0

(2)

Thus, the restriction α1 + α2 + α3 + ... αm = 1 implies constant economies of scale for a given period t. Note that the time-varying framework allows testing if the restriction is fulfilled for other periods. Therefore, we can consider a log-linear production function with variable scale economies as described in Fried, Lovell and Schimdt (1993). Following Battese and Coelli (1995), the inefficiency term uit is assumed to be influenced by p environmental factors:

232

[The Spanish airport system]

uit = ∑o=1 zoitδl + ωit

(3)

The zit represents the p factors affecting the airport’s operational activity that are not managerial controllable or not influenced by management’s decisions (fixed effects). With this in mind, the stochastic production function is not fully flexible. The vit are the random variables assumed to be identically distributed as vit~N(0,σv2) and independent of uit. When an airport is technically efficient (uit = 0), then the error term is symmetric (εit = vit). The random variable (ωit) is defined by the truncation of the normal distribution with zero mean and constant variance (σ2), such that the point of truncation is −zitδ (i.e. ωit ≥ −zitδ). The environmental variables (zit) are to be included within the translog distance function. This is performing one-stage regression since biases are substantial when using two-step procedures (Wang & Schmidt, 2002). In the context of the Spanish airport system, where airports do not compete and significant expansions and investments have taken place without corresponding increases in traffic, the SFA approach is more reliable in terms of individually analysing the impact of the diversity of inputs used and traffic generated in the overall technical inefficiency of the system. It is also effective in demonstrating whether airports’ inefficiencies are consequences of managerial decisions or externalities beyond the central government’s control.

233

THE PUBLIC SPHERE â&#x20AC;&#x2020;| â&#x20AC;&#x2020;2017 ISSUE

DATA DESCRIPTION FIGURE 1 THE SPANISH AIRPORT SYSTEM FROM 2013

Source: AENA.

9. Madrid-Torrejon is used partially as an alternative to MadridBarajas for private civilian flights. From January 2013 all the private flights are transferred to MadridBarajas.

10. Algeciras (Heliport) was opened in July 2010.

234

The Spanish airport-system contains 49 civilian airports, including four general aviation airports (Madrid Cuatro-Vientos, Madrid-Torrejon,9 Sabadell, and Son Bonet) and two heliports (Algeciras and Ceuta). All the airports are government-owned and -managed. The divergence shown in terms of traffic and air traffic control, implying different production functions, have led to the removal of these airports from the sample. The final panel data refers to 43 civilian airports from 2009 to 2013 (215 observations). Data, other than calculated depreciation expenses, has been obtained directly from annual reports of the AENA. Observations begin in 2009, as there is no prior airport-level data available. Table 2 shows the group of airports classified in terms of traffic, expressed as passengers per year. Note that airports are classified in one cluster when showing a consistent number of passengers across the years. By this measure, there are 14 large airports, 13 medium, and 22 small. Table 3 shows the descriptive statistics for the 49 airports.10 There is high variability in terms of passengers and cargo.

[The Spanish airport system]

Table 2 Average airports size (2009-2013) Airports

Size

Min PAX

Max PAX

Alicante; Barcelona; Bilbao; Fuerteventura; Gran Canaria; Ibiza; Lanzarote; Madrid Barajas; Malaga; Palma de Mallorca; Sevilla; Tenerife-North; TenerifeSouth; Valencia

> 3,500,000 (large)

3,654,957

49,866,113

A Coruna; Almeria; Asturias; Girona-Costa Brava; Granada; Jerez; La Palma; Menorca; Murcia; Reus; Santander; Santiago; Vigo

≤ 3,500,000 to > 750,000 (medium)

728,428

5,286,970

Albacete; Algeciras; Badajoz; Burgos; Ceuta; Cordoba; El Hierro; Huesca; La Gomera; Leon; Logrono; Madrid-4-vientos; MadridTorrejon; Melilla; Pamplona; Sabadell; Salamanca; San Sebastian; Son Bonet; Valladolid; Vitoria; Zaragoza

≤ 750,000 (small)

751,097

Source: Author, with passenger data provided by AENA.

The initial depreciation charges published by AENA are highly correlated with the operating costs (0.9915), so new depreciation figures have been estimated based on the construction certifications of works performed in the airports. It has not been possible to find individual infrastructure information before 2000 and after 2011, so calculations are made as if airports start their activity in 2000. In other words, the net book value of the airports’ assets is assumed to be equal to zero in 1999. Airports’ initial investments for 2000 have been estimated by using the depreciation charges for 2004 (accidentally released by the Spanish Government). These depreciation expenses were presented per airport within individual income statements that seemed to reflect a fairness usage of the airports’ assets at that time. The improvements made from 2000 have increased the initial value of the historical costs accordingly. The new depreciation is classified depending on the type of assets:

235

THE PUBLIC SPHERE | 2017 ISSUE

Table 3 Airport summary statistics (2009-2013) Observations

Mean

Standard Dev.

Min.

Max.

Passengers

244

3,960,844

8,543,746

4.99e+07

Air Traffic Movements

244

41,575.27

71,837.34

476

435,187.00

Cargo (t)

244

1.30e+07

5.25e+07

3.94e+08

Commercial Revenues (€)

244

11,756.66

27,875.82

169,507.70

Labour Costs (€)

244

7,256.06

10,202.30

108.12

74,241.79

Operating Costs (€)

244

19,102.79

49,947.44

333.38 318,299.50

Depreciation AENA (€)

244

14,460.75

39,382.69

181.37 264,446.20

Depreciation Airside (€)

244

4,148.64

11,030.28

79,797.00

Depreciation Landside (€)

244

1,086.71

2,208.32

11,941.00

Runway Surface (m2)

244

2,400

927,000

Variable

11. Examples of airside assets are terminals (passengers, cargo and aviation areas), aprons, runways, airfields, control towers, and beacon systems. Landside assets are emergency buildings, parking spaces, and other investments such as electric and power systems, water and recycling systems, etc.

236

171,514.49 161,188.10

Source: AENA except for depreciation airside-landside.

airside assets understood as directly related to the aeronautical activity (operational) and the landside assets that are assumed not to be strictly necessary to develop the airports’ traffic.11 The useful life of the assets is stated according to current regulation in the transportation sector for buildings and structures (1993 to 2005 and from 2006 to date). Following the international financial re-

[The Spanish airport system]

porting standards (IFRS) for property, plant and equipment (PPE, IAS16), any improvement made from 2001 onwards increases the historical cost of the specific asset and is depreciated accordingly at the moment that the asset is ready to be used. For practical purposes the remaining useful life has not been estimated and therefore, the same depreciation percentage has been applied after the improvements. Due to the financial crisis starting in 2008, small- and mediumsized airports have suffered a significant reduction in air traffic compared to large airports. From 2004 to 2007, the vast majority of small and medium airports increased their number of passengers. Nevertheless, from 2007 to 2013, the reduction in traffic was so drastic that only two airports have positive variations (27.86 per cent Santander and 1.12 per cent Santiago). Changes in passengers and air traffic movements are similar across all small and medium airports, except for cargo-specialised Zaragoza, which had significant increases after the set up of a logistical centre (Morales, 2010). Additionally, the geographical distribution of airports shows a significant number of overcrowded areas. A catchment area is defined as the number of airports within one hour and 30 minutes distance driving from the specific airport i (approximately 150 kilometres). The catchment areas include military airportsâ&#x20AC;&#x2122; infrastructure shared with civilian airports and general aviation and heliports used as support to civilian transport. The analysis of catchment areas for 51 airports â&#x20AC;&#x201D; the 49 airports managed by AENA and two airports managed by the Catalonian regional local authority (Lleida and Pirineus-Andorra) â&#x20AC;&#x201D; reveals 35 airports to be within at least one shared influence area. There are 14 airports with another airport in its vicinity; 11 airports with two other airports, and nine airports with three other airports. Vitoria airport can be considered an extreme case, as six other airports lay in its vicinity (Figure 2 on next page). The airports that have three other airports are mostly located in the North of Spain relatively closed to each other.12 This leaves only 16

12. These are two large airports, Barcelona in the Northeast (with an average of 33 million passengers) and Bilbao in the North (4 million); one medium, Reus (Northeast) and four small airports, three in the Northern area (Logrono; Pamplona; San Sebastian) and one in the North-east (Lleida).

237

THE PUBLIC SPHERE â&#x20AC;&#x2020;| â&#x20AC;&#x2020;2017 ISSUE

13. Five big airports with four of them located in the islands: Fuerteventura; Gran Canaria; and Lanzarote (Canary Islands); Ibiza(Balearic Islands) and Valencia. Four medium airports: La Palma (Canary Islands) and Menorca (Baleraric Islands); Almeria and Asturias. And seven small airports: Albacete; Badajoz; Algeciras and Ceuta (Heliports); Melilla and in the Canary Islands, El Hierro and La Gomera.

airports managed by AENA operating on their own within a radius of 150 km (33 per cent).13 FIGURE 2 CATCHMENT AREAS (NORTHERN)

Source: Author.

14. Earnings Before Interest; Taxes; Depreciation and Amortisation.

238

The significant number of airports within the same area, along with excessive investments (leading to high fixed costs) and the centralised decision process wherein airport managers cannot negotiate directly with airlines, results in a loss-making network. From 2009 to 2013, only 19 airports out of 49 achieved positive EBITDAs,14 suggesting that only 39 per cent of airports generated enough revenues through operational activity to cover fixed costs. Table 4 shows that airports with lower level of passengers per year tend to achieve more negative results. This is unsurprising, since EBITDA is an indicator that measures returns based on the ordinary course of business, before subtracting the depreciation

[The Spanish airport system]

and borrowing costs. All profitable airports have a significant number of passengers as well as depreciation charges due to initially extensive infrastructure. These airports are covering their fixed costs through their operational activity (traffic), maintaining a lower ratio of infrastructure per passenger. Nevertheless these observations are not yet confirmed; the depreciation charges published by AENA seem not to be in accordance with how the aeronautical revenues are generated (matching principal). Meetings with airports’ managers have confirmed the aggressive accounting depreciation policies applied by AENA.15 Additionally, the fact that an airport is profitable does not imply that it is technically efficient, as it may have an excess capacity for current and forecasted levels of traffic. The question is whether adequate managerial decisions can be made in order to reduce the technical inefficiencies by attracting more airlines and passengers. Table 4 Average EBITDA (2009-2013) Number of Airports

EBITDA (€)

Depreciation AENA (€)

PAX

Losses until -1 million

-466,000

2,048,000

513,425

-1 to < -2

-1,531,250

2,419,250

365,750

-2 to < -3

-2,544,286

3,257,429

265,632

-3 to < -4

-3,479,600

2,044,000

316,094

-4 to < -5

-4,376,000

2,580,500

345,789

-5 to < -6

-5,664,000

4,373,333

424,126

< -7

-9,984,000

3,664,000

28,304

EBITDA > 0

51,999,095

36,610,421

9,642,768

15. Granada is a medium airport with losses between three and four million euros; Jerez with losses between four and five million euros and La Palma with more than five million euros. All these airports, being medium, have significant depreciation charges exceeding their negative EBIDTAs (based on AENA’s depreciation).

Source: AENA.

239

THE PUBLIC SPHERE | 2017 ISSUE

Taking into account data availability, reliability and conventions from the literature, three inputs and four outputs were selected for inclusion in the SFA model. The outputs are the annual number of passengers (PAX), air traffic movements (ATM), cargo, and commercial revenues. Inputs are labour costs (excluding air traffic control services), operating costs, and the depreciation of airside assets. Aeronautical revenues are not included since these are contained in the total value of passengers, air movements and cargo. In order to avoid convergence problems, the data was deflated according to the Spanish GDP deflator (base 2005). The variables were standardised by the respective geometric mean (Fleming & Wallace, 1986). This allows for the estimation elasticities at the sample means (Cuesta and Orea, 2002).

ESTIMATION OF THE TRANSLOG DISTANCE FUNCTION The translog distance function (1) for the three inputs and the four outputs for each airport i and period t, arbitrarily choosing the labour costs as the input for normalisation purposes, is seen on the following page. This function provides some insight regarding the interaction of inputs and outputs with fixed costs – in this case, labour costs. It is therefore a reflection of the effectiveness of management, which is currently centralised in the AENA. One would expect efficient airports to generate enough aeronautical revenues to overcome fixed costs, in this case, labour costs. One would also expect to find a relationship between the usage of an airport’s infrastructure and the number of employees working at the airport. The interaction between inputs and outputs will reflect their overall impact on staff costs. For example, it would be expected that more passengers will also imply more traffic movements (busy airports) or even more cargo. But at the same time, large

240

[The Spanish airport system]

FIGURE 3 TRANSLOG DISTANCE FUNCTION

ln(Labour ) it

= β0 +

Operatingit β1ln( Labour ) it

+ 1/2β'2ln(

Depreciation Assetsit β2ln( ) Labourit

Depreciation Assetsit )2 Labourit

+ 1/2β12ln(

+ 1/2β'1ln(

Operatingit 2 Labourit )

Operatingit Depreciation Assetsit ) Labourit Labourit )ln(

+ β3ln(Commercial Revenuesit) + β4ln(PAXit) + β5ln(ATMit) + β6ln(Cargoit) + 1/2β'3 (lnCommercial Revenuesit)2 + 1/2β'4 (lnPAXit)2 + 1/2β'5 (lnATMit) 2 + 1/2β'6 (lnCargoit) 2 + 1/2β34 ln(Commercial Revenuesit) ln(PAXit) + 1/2β35 ln(Commercial Revenuesit) ln(ATMit) + 1/2β36ln(Commercial Revenuesit) ln(Cargoit) + 1/2β45ln(PAXit) ln(ATMit) + 1/2β46ln(PAXit) ln(Cargoit) + 1/2β56ln(ATMit) ln(Cargoit) + β31ln(Commercial Revenuesit) ln( + β51ln(ATMit) ln(

Operatingit Labourit )

+ β61ln(Cargoit) ln(

+ β31ln(Commercial Revenuesit) ln( + β42ln(PAXit) ln(

Depreciation Assetsit ) Labourit

+ β62ln(Cargoit) ln(

+ β41ln(PAXit) ln(

Operatingit Labourit )

Depreciation Assetsit ) Labourit

+ β52ln(ATMit) ln(

Depreciation Assetsit ) Labourit

+ (vit- uit)

Above: Translog distance function (1) for three inputs and the four outputs for each airport i and period t, arbitrarily choosing the labour costs as the input for normalisation purposes.

airports in terms of traffic will have to cover higher depreciation expenses, potentially having a negative impact in overall labour costs. I return to these interactions when I present results in the next section.

241

THE PUBLIC SPHERE | 2017 ISSUE

The component uiof the error in (3) in terms of the specific environmental variables is modelled:

ui = δ0 + δ1 (Catchment Area)+δ2 (PSO)+δ5 (Accessibility)+δ6 (Capacity Utilisation)+δ7 (Type of Airport)+ωit While environmental factors may be affecting airport efficiency, manipulating these is not feasible, at least in the short-term. PSO is a dichotomous variable that equals one if an airport has public service obligation routes, and zero otherwise. The Spanish airport system has obligatory routes in the intra-insular flights (in the Canary and the Balearic Islands) and, starting in April 2009, from Almeria to Sevilla. These routes are ensured by the Spanish government in areas identified as remote (except the latest one). Accessibility is a dummy variable that equals one if there are train facilities for direct access to the airport. There are currently five airports with railway infrastructure: four big airports (Alicante, Barcelona, Madrid and Malaga) and one medium (Jerez). Capacity Utilisation refers to the available capacity of an airport in light of air traffic control restrictions. This variable has three categories: airports with a low level of coordination; airports partially coordinated; and airports fully coordinated (slots). The first category is considered to satisfy the current and potential demand of airlines; these are normally small airports with a surplus capacity due to low traffic demand. The second group represents airports restricted during the summer and those with schedules imposed during the whole year. These are usually medium-sized airports with significant seasonal effects (peaks of demand during tourist seasons). Their capacity is usually close to their actual demand and other airports located at close distance may be used in congestion periods. Finally, the slots are large air-

242

[The Spanish airport system]

ports with high demand and with a significantly lower capacity compared to their current and forecast demands. These are busy airports that are highly restricted in terms of landing and taking off compared to other airports.16 Type of Airport refers to airports’ main operational activity. This variable has the following categories: fully civilian; military air bases partially used as support for civilian transport (Murcia); and army infrastructure used partially or fully for civilian air transport. Note that all environmental variables except catchment areas have been defined as dummies. Consequently, the n−1 cases for each dummy are included in the translog stochastic distance function (in the inefficiency term) without requiring it to be previously log. As stated, the δs are unknown scalar quantities to be estimated. The coefficient δ0 represents the level of inefficiency of a specific airport without considering the existence of uncontrolled external factors. The rest of the coefficients (δ1) are externalities representing the individual effect of the respective environmental variable explaining the technical inefficiency. The variance parameter provides insight regarding the explanatory power of the model (i.e. the inputs and outputs used) compared to the variability that is not possible to be explained (noise). A good model will have a low-value of the noise variance (σv2) compared to the inefficiency term (σu2) expressed in terms of lambda (λ)17:

σ2 = σu2 + σv2

σu σ λ = 2√ σ2 = σ v v 2 u

(4) (5)

A high value of lambda (λ) implies that technical inefficiency of the airports is likely to significantly influence the value of the

16. The current slots are Alicante; Barcelona; Bilbao; Madrid; Malaga; Valencia; in the Canary Islands: Fuerteventura; Gran Canaria; Lanzarote; Tenerife South and in the Balearic Islands: Palma de Mallorca.

17. This is equivalent to the γ-gamma coefficient in Battese and Coelli (1988). In this model gamma indicates how much of the variance of the composed error term is attributed to the technical inefficiencyterm (γ=σu2/σv2).The rejection of the null hypothesis H0: γ=0 implies the existence of a stochastic production frontier.

243

THE PUBLIC SPHERE | 2017 ISSUE

dependent variable, in this case, the labour costs. Low lambda values are signals for a significant weight of noise compared to the explained inefficiency: The inputs and outputs used are not relevant for determining the dependent variable and the inefficiency term. Consequently, the potential inefficiency would not be explained either by the way that airports are managed or by externalities.

RESULTS The results are summarized in Table 5. The first column (SFA1) corresponds to the translog model without considering the inclusion of environmental variables. This model corresponds to Battese and Coelli (1992) and assumes that all inefficiencies stem from airport management (AENA). The second column corresponds to the same model, but includes the environmental factors in one stage (Battese and Coelli, 1995). This model argues that not all inefficiencies can be reduced, since there are external factors affecting the airports’ traffic beyond managerial control. Note that only the significant iterations are shown in the production side (βs), but all environmental variables are enclosed in the inefficiency term for comparison purposes (δs). Both models have significant explanatory power with low presence of noise (high values of log-likelihood; significant and high lambdas). The expected value of technical inefficiency is significantly different from zero (μ = 20.86 per cent), allowing for environmental variables (SFA2). The results obtained in the translog stochastic distance function are similar to Coelli, Perelman and Romano (1999) since the environmental variables are influencing air transport; neglecting them would bias the estimation of the model (Hattori, 2002). This approach recognises the existence of externalities which contribute to airports’ inefficiencies but are unchangeable in the short-term. Both models show consistency

244

[The Spanish airport system]

Table 5 Translog alternative distance functions (43 airports 2009-2013) Coefficients

SFA1

SFA2

β0

Constant

.5774*

.4703*

β3

LnCommercial Revenues

-.2232*

-.1896*

β4

LnPAX

-.0294

-.0937*

β5

LnATM

-.142*

-.1387*

β6

LnCargo

-.0609*

-.0493*

β1

LnOperating Costs

.4364*

.4595*

β2

LnDepreciation Assets

.0005

β'1

½LnOperating Costs²

-.2367*

-.1857¹

β'2

½LnDepreciation Assets²

-.0002

-.0004*

β34

½LnCommercial Revenues∙LnPAX

-.1634*

-.1873*

β35

½LnCommercial Revenues∙LnATM

.2336*

.203¹

β36

½LnCommercial Revenues∙LnCargo

-.0235*

-.021

β46

½LnPAX∙LnCargo

.0251*

.0296* -.0248

β56

½LnATM∙LnCargo

-.0488*

β12

½LnOperating Costs∙LnDepreciation Assets

.0102*

.0161*

β51

LnATM∙LnOperating Costs

-.0812

-.1404*

β32

LnCommercial Revenues∙LnDepreciation Assets

-.0064*

-.0047*

β62

LnCargo∙LnDepreciation Assets

.0014*

.0015*

�

Mu (Inefficiency)

.2086*

δ1

Catchment Area

-.1977

δ2

PSO Routes

-.2839

δ3

Accessibility

1.948

δ4

Partially Coordinated

5.144

δ5

Low Coordinated

5.699*

δ6

Civilian-Military

-9.812

δ0

Constant

σu

Sigma - u

.2065*

.6441*

σv

Sigma - v

.0210

.1306*

Lambda (σu/σv)

9.812*

4.930*

Log Likelihood

94.86

91.92

*Significantly different from zero at least at 5 per cent. 1

SFA2: 1⁄2LnOperating Costs2 P > |z| = 0.098; 1⁄2LnCommercialRevenues∙LnATM

P > |z| = 0.086. Source: Author.

245

THE PUBLIC SPHERE | 2017 ISSUE

in the results obtained for the basic linear variables. The signs of the simple effects are also as expected: Outputs decrease the labour costs and inputs increase them. Commercial revenue is the most significant output financing the labour costs (β3= −22 per cent ; −19 per cent). The relevance of commercial revenues is confirmed by the low, but significant, impact of the number of passengers in the second model (β4 = −9 per cent). The literature shows that the diversification from aeronautical activities towards commercialisation corresponds to airports’ privatisation process (Humphreys, 1999). In the Spanish case though, all the airports continue to be public utilities. Therefore, these results confirm the overall inability of airports to generate enough aeronautical revenues to cover the operating costs (confirmed by the EBIDTA analysis). This is a reflection of an inadequate decision-making process and a need to diversify income sources. The fact that individual airport managers cannot make decisions with regard to resource allocation or levels of production results in airports being underutilised. Individual managers are familiar with daily operations, regional needs and local demand. Consequently, it seems sensible for managers to have flexibility and be involved in the decision-making process. The fact that the depreciation is not significant (β2) – or, being significant, has very low relevance (β'2 = −0.04 per cent) – the low impact of the number of passengers (β4) and overall airport orientation towards commercialisation (β3) calls for a critical assessment. Responsiveness to these measures of basic operations helps us understand whether Spanish airports are really used for aeronautical purposes and commercial activities, or whether they may be suffering from over-capacity. With this in mind, the simple effect of passengers cannot be analysed separately from commercial revenues. As expected, the interaction between commercial revenues and passengers is high and significant (β34 = −16 per cent; −19 per cent). The major impact of privatisation is on commercial revenues, not infrastructure, as this is generally

246

[The Spanish airport system]

already present. These results show how commercial revenues are a significant source of the overall airports’ income: Commercial revenues are as important as aeronautical revenues (ICAO, 2013). Operating cost is the input with the largest negative impact on labour costs (β1 = +44 per cent;+46 per cent). These results seem sensible due to operating costs being variable in nature: Airports with more traffic will experience higher operating costs, hence require more staff. The results also show that when airports achieve a certain level of traffic, the operating costs’ effect reverses (β'1 = −24 per cent; −18 per cent). Operational efficiency is greater with increasing scale. The question remains whether this effect is general or only for airports with a certain level of traffic; unless enough traffic is generated, some of the smallest airports will contribute to the modelled system by closing, avoiding costs. The analysis of catchment areas becomes a guide for deciding which airports should be closed. Such airports are usually not efficient from an economic perspective (EBIDTAs) rather than technically inefficient; they contribute more to the system if they do not remain opened. This is confirmed by the significance of the constant being significant and showing high values in both models (β0 = +58 per cent; +47 per cent). Airports without traffic will translate into labour costs equal to the value of the constant. Spanish airports have permanent staff independent of their traffic, but related to their investments to ensure the maintenance of infrastructure and installations. The high value of the constant highlights that the government may be considering social welfare before actual industry needs (airports remaining open for connectivity purposes or to avoid dropping the employment rate). Although AENA publicly announced the consideration of closing nine airports serving less than 100,000 passengers, the Spanish government seems committed to keeping all Spanish airports open, even at a loss.18 None of the environmental variables are significant, nor are the number of airports within the same influence area (δ1). This

18. Airports with less than 100,000 passengers will lose 130 euros per passenger and year. These airports are not financially selfsustainable and will struggle to remain in operation without more public money (European Court of Auditors, December 2014).

247

THE PUBLIC SPHERE | 2017 ISSUE

is supported by the fact that centralised management regarding commercial policies undermines competition. The existence of public service obligation routes (δ2) also does not explain airports’ inefficiencies. Finally, the fact that airports are sharing their infrastructure with military activities does not influence the inefficiency of airports (δ6). Spanish airports suffer from technical inefficiencies and high fixed costs (labour and depreciation). The analysis of environmental variables along with the production function concludes that there are no externalities causing the airports’ inefficiencies; rather, management decisions made regarding inputs and outputs are to blame. Airports are becoming more commercially oriented since the airports’ managers do not have decision-making power over other revenue-generating activities. The centralised decisions regarding commercial policies do not make the Spanish market attractive to airlines and passengers. The system contains a significant number of small and medium airports operating within the same area, but these are not allowed to compete. Expansions and investments in infrastructure are not required, and have resulted in a non-profitable airport system due to over-capacity. This inefficiency – infrastructure not used for aeronautical purposes – has also led to an increasing public deficit and the allocation of financial resources towards unprofitable airports.

CONCLUSIONS 19. Overall only 13 airports out of 49 (12 big airports and one medium, Girona) have shown consistent positive operating results during the years of the study. All the small and medium airports (except Girona) are not- profitable.

248

Most of the airports are not profitable as a consequence of low levels of traffic. Consequently, the unprofitable airports have idle infrastructure. Airports with over-capacity cannot overcome their fixed costs. These airports become a burden for the system since they are cross-subsidised by profitable airports.19 The fact that the main aeronautical outputs, such as passengers and cargo, are not significant (or with very low impact) clearly evidences a

[The Spanish airport system]

network with airports underused. Airports are forced to diversify their main operational course of business from aeronautical activities towards commercialisation. Although commercialisation has been evidenced when airports are privatised, in the Spanish airport system the commercial revenues have become the main source of income rather than air transport within a public system. Large airports in terms of passengers are profitable, with a better initial use of their infrastructure compared to medium and small airports. Large airports are not only profitable, but also efficient. Therefore, the question is whether the small and medium airports are strictly necessary for connectivity purposes. The analysis of catchment areas shows regional areas overloaded with small and medium airports. Additionally, the lack of flexibility in terms of management leads to the absence of competition, as shown by the catchment areasâ&#x20AC;&#x2122; saturation being irrelevant for efficiency. The discussion is if these airports could be closed and their traffic transferred to the most technically efficient alternative, but with a minimal impact on connectivity. The recommendations of the European Court of Auditors (December 2014) have not been followed. It is not clear why the Spanish government is determined to keep all airports open, even when traffic is lacking considerably: these airports bleed the public purse. The critical aspect for individual airport performance relies on the managerial decisions regarding the inputs negatively affecting airportsâ&#x20AC;&#x2122; technical efficiencies. This implies that decisions among inputs could be made to reduce inefficiencies. AENA should reconsider enhancing the aeronautical aspect of the airports. Airports with more infrastructure, but lower levels of traffic, will definitely become more inefficient unless competition is enhanced by avoiding standardization of the decision making process. This implies allowing the individual airportsâ&#x20AC;&#x2122; managers to freely decide commercial policies in order to ensure competitive prices and quality of the services provided, making the Spanish airports more attractive to airlines and passengers.

249

THE PUBLIC SPHERE | 2017 ISSUE

Following the conclusions of the Audit report from the European Commission (December 2014), this study is a critical assessment of the impact that centralised management has had on the Spanish airport system’s efficiency. The international financial reporting standards (IFRSs) require that financial statements faithfully represent reality. Airports’ privatisation, commercialisation, infrastructure, market regulation and growth of traffic are a source of dynamism and uncertainty. The centralisation of decision-making cannot survive in a changing environment. Accounting policies cannot be applied to all airports equally, as airports use their infrastructure to generate revenue differently from one another. Centralised decisions, without consideration for the specificities of each airport and their infrastructure use, potentially breaches the matching principle and accruals convention. Some critical matters, such as aggressive policies regarding depreciation; aeronautical revenues not correlated to traffic, but based on discriminatory increases of air fares; commercial policies decided at governmental level without flexibility granted to airports’ managers in negotiating with airlines; and a significant number of small and medium airports located within the same areas and not competing; calls for an assessment of the Spanish airport system regarding profitability and management efficiency. Results show that most of the inefficiencies are due to restricted managerial decisions (inputs and outputs) rather than negative externalities. The government ownership-management scheme results in a system which is costly to run, with low aeronautical income and significant debt incurred to finance investments not required for actual and forecast demand. Airport managers require flexibility deciding commercial policies to diversify in terms of price and type of services (CNMC, 2014), but is decentralisation and deregulation in the Spanish airport industry possible? Unless the European Commission starts requiring from AENA transparent and individual financial statements, including the relevant disclosures, it is unlikely to happen.

250

[The Spanish airport system]

REFERENCES Battese, G.E. & Coelli, T. J. (1988): ‘Prediction of Firm-level technical efficiencies with a generalized frontier production function and panel data’ Journal of Econometrics 38, 387- 399 Battese, G.E. & Coelli, T. J. (1992): ‘Frontier production functions, technical efficiency and panel data: with application to paddy farmers in India’. Journal of Productivity Analysis 3, 153-169 Battese, G.E. & Coelli, T. J. (1993): ‘A Stochastic Frontier Production Function incorporating a model for technical inefficiency effects’. Working Papers in Econometrics and Applied Statistics 69, Department of Econometrics. University of New England. Battese, G.E. & Coelli, T. J. (1995): ‘A Model for Technical Inefficiency Effects in a Stochastic Frontier Production Function for Panel Data’. Empirical Economics 20, 325-332 Charnes, A.; Cooper, W. W.; Rhodes, E. (1978): ‘Measuring the efficiency of decision making units’. European Journal of Operational Research 2, 429-444 Coelli, T. J. & Perelman, S. (1996): ‘Efficiency measurement, multi-output technologies and distance function: with application to European railways’, CREPP Working Paper 96/05, University of Liege. Coelli, T. J.; Perelman, S. & Romano, E. (1999): ‘Accounting for Environmental Influences in Stochastic Frontier Models: with application to International Airlines’ Journal of Productivity Analysis 11, 251-273 Coelli, T. J.; Prasada Rao, D.S.; O’Donnell, C.J. & Battese, G.E. (1998): ‘An Introduction to Efficiency and Productivity Analysis’. Kluwer Academic Publishers, New York. Costas-Centivany, C.M. (1999): ‘Spain’s airport infrastructure: adaptations to liberalization and privatisation’ Journal of Transport Geography 7, 215-223 Coto-Millán, P.; Carrera-Gómez, G.; Castanedo-Galán, J.; Inglada, V.; Núñez-Sánchez, R.; Pesquera, M.A.; Sainz, R. (2007): ‘Efficiency Stochastic Frontiers: a Panel Data Analysis for Spanish Airports (1992–1994)’ Essays on Transport Economics (chapter 121-126), Physica-Verlag HD Coto-Millán, P.; Casares-Hontañón, P.; Inglada, V.; Agüeros, M.; Pesquera, M.A.; Badiola, A. (2014): ‘Small is beautiful? The impact of economic crisis, low cost carriers, and size on efficiency in Spanish airports (2009–2011)’ Journal of Air Transport Management 40, 34-41 Coto-Millán, P.; Inglada, V.; Fernández, X.L.; Inglada-Pérez, L. & Pesquera, M.A. (2016): ‘The effect procargo on technical and scale efficiency at airports: The case of Spanish airports (2009–2011)’ Utilities Policy 39, 29-35 Cuesta A. & Orea, L. (2002): ‘Mergers and technical efficiency in Spanish saving banks: A stochastic distance function approach’ Journal of Banking & Finance 26 (12), 22312247 Cuesta, R. A., Lovell, C. A. K., Zofío, J. L., (2009): ‘Environmental efficiency measurement with translog distance functions: A parametric approach’. Ecological Economics 68, 2232 – 2242. Färe, R.; Grosskopf, S.; Lovell, C. A. K. & Yaisawarng, S. (1993), ‘Derivation of Shadow Prices of Undesirable Outputs: A Distance Function Approach’. Review of Economics and Statistics 75(2), 374-380. Fried, H. O., Lovell, C. A. K. & Schimdt, S. S. (1993): ‘The Measurement of Productive Efficiency’. Cambridge University Press, Cambridge

251

THE PUBLIC SPHERE | 2017 ISSUE

Fleming, P.J. & Wallace, J.J. (1986): ‘How not to lie with Statistics: the correct way to summarize benchmark results’. Communications of the Association for Computing Machinery ACM, 29(3) 218-221 Hattory, T. (2002): ‘Relative Performance of U.S. and Japanese Electricity Distribution: an application of Stochastic Frontier Analysis’. Journal of Productivity Analysis 18, 269284 Humphreys, I. (1999): ‘Privatisation and commercialisation Changes in UK airport ownerships patterns’. Journal of Transport Geography 7, 121-134 Kumbhakar, S.C. (1990): ‘Production frontiers, panel data and time varying technical inefficiency’. Journal of Econometrics, 46 201-2012 Kumbhakar, S.C.; Ghosh, S. & McGuckin, J.T. (1991): ‘A Generalized Production Frontier Approach for Estimating Determinants of Inefficiency in U.S. Frontier Approach for Estimating Determinants of Inefficiency in U.S. Dairy Firms’. Journal of Business and Economic Statistics 9, 279-286 Kumbhakar, S.C.; Lien, G. & Hardaker J.B. (2014): ‘Technical efficiency in competing panel data models: a study Norwegian gran farming’. Journal of Productivity Analysis 41 (2), 321-337 Kumbhakar, S.C. & Lovell, C.A. (2000): ‘Stochastic Frontier Analysis’. Cambridge University Press, Cambridge. Lovell, C. A. K. ; Richardson, S.; Travers, P. & Wood, L. L. (1994) : ‘Resources and Functionings: A New View of Inequality in Australia’. In W. Eichhorn (ed.), ‘Models and Measurement of Welfare and Inequality’. Berlin Springer-Verlag, 787–807 Lozano S. & Gutiérrez, E. (2011): ‘Slacks-based measure of efficiency of airports with airplanes delays as undesirable outputs’. Journal Computers and Operations Research 38(1), 131-139 Martín J.C. & Román, C. (2001): ‘An application of DEA to measure the efficiency of Spanish airports prior to privatization’. Journal of Air Transport Management 7(3), 149-157 Martín, J. C. & Román, C. (2006): ‘A benchmarking analysis of Spanish commercial airports. A comparison between SMOP and DEA ranking methods’. Networks and Spatial Economics 6(2), 111-134 Martín J.C.; Román, C. & Voltes-Dorta, A. (2009): ‘A stochastic frontier analysis to estimate the relative efficiency of the Spanish airports’. Journal of Productivity Analysis 31(3), 163-176 Martín J.C.; Román, C. & Voltes-Dorta, A. (2011): ‘Scale economies and marginal costs in Spanish airports’. Transportation Research Part E 47, 238-248 Martín-Cejas, R.R. (2002): ‘An approximation to the productive efficiency of the Spanish airports network through a deterministic cost frontier’. Journal of Air Transport Management 8 (4), 233–238 Morales Gil, A. (2010): ‘Las superficies logísticas y la organización espacial de redes de transporte de mercaderías en España’. Papeles de Geografía 51-52, 211-222 Murillo-Melchor, C. (1999): ‘An Analysis of technical efficiency and productivity changes in Spanish Airports using the Malmquist Index’. International Journal of Transport Economics 26(2), 271-292 Porter, M.E. (1979). ‘How Competitive Forces Shape Strategy’. Harvard Business Review, March-April Salazar de la Cruz, F. (1999): ‘Technical, allocative and economic efficiency in transport’. International Journal of Transport Economics 26(2), 255-270

252

[The Spanish airport system]

Schmidt, P. & Sickles, R. (1984): ‘Production frontiers and panel data’. Journal of Business and Economics Statistics 2 (4), 367-374 Simar, L. & Wilson, P.W. (1999): ‘Of course we can bootstrap DEA scores! But does it mean anything? Logic trumps wishful thinking’. Journal of Productivity Analysis 11, 93-97 Simar, L. & Wilson, P.W. (2007): ‘Estimation and inference in two stage, semi-parametric models of productive efficiency’. Journal of Econometrics 136, 31-64 Tapiador, F.J.; Mateos, A. & Martí-Henneberg, J. (2008): ‘The geographical efficiency of Spain’s regional airports: a quantitative analysis’. Journal of Air Transport Management 14(4), 205-212 Tovar, B. & Martín-Cejas, R.R. (2009): ‘Are outsourcing and non-aeronautical revenues important drivers in the efficiency of Spanish airports?’ Journal of Air Transport Management 15(5), 217-220 Tovar, B. & Martín-Cejas, R.R. (2010): ‘Technical efficiency and productivity changes in Spanish airports: A parametric distance functions approach’. Transportation Research Part E 46(5), 249-260 Wang, H. & Schmidt, P. (2002): ‘One-Step and Two-Step Estimation of the Effects of Exogenous Variables on Technical Efficiency Levels’. Journal of Productivity Analysis 18(2), 129-144. Reports Abertis (2009): ‘Ten years that marked an historic era change in the presidency’. Issue 17 ACETA (2011): ‘Estudio y Analisis sobre el incremento de las tasas aeroportuarias en aviacion general y en la formacion de pilotos profesionales’ http://aviaciondigital.com/ incremento-de-tasas-aeroportuarias-en-aviacion-general-y- formacion-de-pilotos/ http://www.aviaciondigitalglobal.com/newsFiles/20110622011312-4.pdf ACI Europe (2010): ‘The Ownership of Europe’s Airports’ European Commission European Court of Auditors (2014): ‘EU-funded airport infrastructures: poor value for money’ Special Report, 21 http://www.politico.eu/article/euwasted-money-on-new-airports-say-auditors/ European Commission Press release (July, 2016): ‘State aid: Commission orders Spain to recover incompatible State aid for high speed train test centre from railway operator ADIF’ http://europa.eu/rapid/press-release_IP-16-2621_en.htm Cambra de Comerç de Barcelona (November 2010): ‘El modelo de gestión aeroportuaria en España: marco institucional y jurídico y líneas maestras para una propuesta de cambio’ Comisión Nacional de los Mercados y la Competencia (CNMC, July 2014): ‘El Sector Aeorportuario en España: Situación Actual y Recomendaciones de Liberización’ International Civil Aviation Organization (ICAO) (March 2013): ‘Worldwide Air Transport Conference (ATCONF) Sixth meeting: Airport Competition’ https://www.google.com/ maps/d/viewer?mid=zylWTnxRlUJE.kU5cVv1m27wk&hl=en_US Word Finance (March, 2016): ‘Spain’s infrastructure woes are more serious than they appear’ http://www.worldfinance.com/infrastructure-investment/government-policy/ spains- infrastructure-woes-are-more-serious-than-they-appear

Please see publicspherejournal.com for online appendices and acknowledgements.

253

Monetary policy at the zero lower bound Discretion, uncertainty and interest rate volatility Mark A.C. Wilkinson

Master of Public Administration, Class of 2016 London School of Economics and Political Science

ÂŠAshley Lau

ABSTRACT This study investigates optimal monetary policy under a set of simultaneous constraints: i) when nominal interest rates are at their zero lower bound (ZLB), ii) when there is uncertainty about the natural interest rate, and iii) when the central bank cannot make credible commitments. In these situations, there are asymmetric risks in setting monetary policy, as the ZLB prevents the central bank from further decreasing nominal rates to stimulate the economy, and unconventional monetary policy is an imperfect substitute for the constrained traditional interest rate instrument. There is therefore a trade-off between keeping interest rates lower for longer, to ensure recovery, and beginning to raise rates earlier but more gradually. To assess the ability of different policies to handle this trade-off, I use a New Keynesian dynamic stochastic general equilibrium model with monetary policy defined either by a standard Taylor rule or discretionary inflation targeting that considers uncertainty. The central finding is that an optimal discretionary policy augmented to account for uncertainty about the natural interest rate forcing the ZLB to bind can perform as well or better than the Taylor rule by staying lower for longer, even though the discretionary policy has a more volatile policy rate. A further result is that a Rogoff-type conservative central banker who puts a high relative weight on stabilizing inflation can alleviate the deflationary pressure of the ZLB episode.

[Monetary policy at the zero lower bound ]

INTRODUCTION A fundamental constraint faced by central banks is that they cannot make fully credible commitments to future policy actions. The resulting degree of discretion has been the status quo at central banks, but in recent years this has been complicated by two conditions: the zero lower bound (ZLB) on nominal interest rates and uncertainty about future economic activity. The objective of this study is to specify the characteristics of optimal monetary policy under these conditions in a standard New Keynesian model extended to incorporate uncertainty in the central bank’s projection of the natural interest rate. The natural interest rate is a proxy for economic activity because it is the real interest rate that is consistent with stable prices and full employment. The central problem of the ZLB is that interest rates below zero entail making lenders pay to hold their money in a bank. In this situation, the rational lender would prefer to hold their money in paper currency, thereby earning zero per cent interest instead of having to pay the bank. The availability of paper currency as an alternative investment acts as an interest rate floor and prevents central banks from pushing the policy rate—the overnight nominal interest rate at which banks lend to each other—significantly below zero. This dynamic is known as a liquidity trap, since the conventional policy of expanding liquidity through monetary easing becomes impotent (Krugman, 1998; Eggertsson, 2006; Svensson, 2006).1 In weak economic times, the central bank’s primary tool is to lower the policy rate to stimulate output and inflation. As the policy rate approaches zero, the ZLB hinders economic recovery by preventing central banks from decreasing it any further. This barrier makes it particularly difficult to combat persistently low inflation, and the prospect of interest rates at the ZLB can even induce deflation—a widespread decline in prices. The ZLB has proved a serious constraint for the Bank of Japan since the late

1. A stricter definition is that an economy is not in a liquidity trap until the central bank has bought enough of the economy’s interest-bearing assets to push all of their interest rates to zero. As put by Orphanides (2003, p. 19), “additional monetary expansion continues to have some bite because the prices and yields of all assets, not merely ‘the’ shortterm nominal rate of interest, jointly determine aggregate demand.” For the purposes of this paper, I maintain the definition associated with the central bank policy rate being at the ZLB.

257

THE PUBLIC SPHERE â&#x20AC;&#x2020;| â&#x20AC;&#x2020;2017 ISSUE

1990s and for many other central banks since the financial crisis of 2007-2008, notably including the Federal Reserve and the European Central Bank. Central banks have employed unconventional monetary policy as an alternative means of affecting interest rates when the policy rate is at the ZLB; these unconventional policies include large-scale asset purchases (LSAPs), forward guidance, and slightly negative policy rates. However, the ZLB remains a serious constraint because these tools are imperfect substitutes for the traditional policy rate. Firstly, the benefits of LSAPs for the wider economy are very difficult to estimate, and there are a series of associated costs that limit the extent of feasible asset purchases (Bernanke, 2012). These include risks to financial instability, the impairment of certain asset markets, beggar-thy-neighbour currency devaluations, and equity concerns. Secondly, negative policy rates, though possible, cannot go below approximately negative 0.5 per cent, as the costs to holding capital as paper currency are not infinite. Therefore, at some point agents will simply choose to invest in vaults, rather than pay exceedingly negative interest rates (Bank of Canada, 2015). Lastly, forward guidance about intended policy actions is not a perfect substitute for actual policy rate changes because central banks are unable to commit credibly beyond the short term. This inability stems from the fact that there are changes in economic conditions, our understanding of the economy, and central bank leaders (Woodford, 2003). The result of the ZLB hindering the first-best policy tool is an asymmetry in policy risks in the central bankâ&#x20AC;&#x2122;s decision-making process. The relative inadequacy of unconventional policy responses means that central banks should give higher consideration to downside risks such as deflation and recession near the ZLB, even if the expected values of the upside and downside risks are the same. Upside risks, such as unexpected increases in output and inflation, can be countered more easily with a rapid

258

[Monetary policy at the zero lower bound ]

increase of the policy rate to an appropriate levelâ&#x20AC;&#x201D;a well-practiced policy action. Assuming that central banks can commit to future actions via forward guidance is instructive for studying the theoretically optimal policy, but this assumption must be relaxed to accord with reality. A discretionary central bank has much more difficulty combating the ZLB because it cannot commit itself to a post-ZLB burst of inflation (Krugman, 1998). Theories of optimal monetary policy assume central banks can fully commit to certain policies and that there is no uncertainty in the projections of key variables by the central bank.2 Furthermore, some studies of these theories do not include a non-negativity constraint representing the ZLB (Rotemberg and Woodford, 1997; Schmitt-Grohe and Uribe, 2004). While normally there are higher welfare losses under discretion relative to commitment, these losses are even higher at the ZLB (Adam and Billi, 2005). Evans et al. (2015) model a discretionary central bank facing the explicit ZLB constraint and uncertainty about whether the natural rate of interest forces the ZLB to bind. Their result is that the optimal discretionary policy postpones raising the policy rate from the ZLB by two to three quarters, and then raises it more steeply, all relative to policies that ignore uncertainty. A counter-argument to this relatively delayed rate increase is that the sudden changes in the policy rate cause social welfare losses through volatility in financial markets (Stein and Sunderam, 2015). I extend the Evans et al. (2015) model to assess this tradeoff, and I focus on discretionary policy at the ZLB because the commitment case is well studied and because the discretionary case is a better depiction of reality for most central banks. First, I augment the welfare criterion, used to judge the desirability of alternative policies, to assess whether a policy rate path that remains â&#x20AC;&#x153;lower for longerâ&#x20AC;? is excessively volatile when it finally leaves the ZLB. The main finding is that optimal discretionary policy does not impose undue welfare losses through pol-

2. See Woodford (2003) for an authoritative review of both optimal commitment and discretion theory.

259

THE PUBLIC SPHERE â&#x20AC;&#x2020;| â&#x20AC;&#x2020;2017 ISSUE

icy-rate volatility, unless society both puts a disproportionately high value on stabilizing the policy rate and cannot hedge against such volatility. Second, given the deflationary bias of purely discretionary policy at the ZLB, I introduce a so-called Rogoff conservative central banker that places a much higher value on inflation stabilization than output gap stabilization. The Rogoff conservative central banker is helpful at increasing inflation back to target during a liquidity trap, with the caveat that there are is high policy rate volatility.

THEORETICAL FRAMEWORK The liquidity trap as a credibility problem During a liquidity trap, the nominal policy rate cannot be decreased to match decreases in the natural rate into negative territory. The ZLB technically binds when the natural rate is so negative that it is lower than the negative of the inflation target; that is, when the natural rate is less than the lowest possible real interest rate (Woodford 2003). The resulting gap of the real rate above the natural rate slows the economy and decreases inflation. A vicious cycle then begins, as decreases in inflation increase the real rate still further above the natural rate. Krugman (1998) articulated the crux of the liquidity trap as a credibility problem: for an expansion of the money supply to raise prices in equal proportion, the central bank must commit to the policy credibly enough to convince the market that the volume of the expansion will not be reversed in the future. The market, however, expects the discretionary central bank to renege and prevent the inflationary burst as soon as the ZLB is no longer binding. Commitment to this post-ZLB boom would, if credible, increase welfare because it would cause lower long-term real interest rates even while the policy rate is stuck at zero. That

260

[Monetary policy at the zero lower bound ]

effect would feed back into a reduced incentive to save during the liquidity trap, thereby spurring output and preventing price cuts in a “virtuous circle” (Woodford, 2010, p. 753). Nevertheless, for discretionary central banks this promise is not credible, and the welfare loss from discretion relative to commitment is even higher at the ZLB (Adam and Billi, 2005).

The canonical New Keynesian model The theoretical framework I use to analyse the optimal monetary policy problem is the canonical New Keynesian stickyprice model considered by Clarida, Galí, and Gertler (1999) and Woodford (2003), inter alia. This framework includes forward-looking behaviour and is derived from micro-foundations of optimization by households and firms.3 The temporary nominal rigidities arise from price-setting behaviour as outlined by Calvo (1983). I begin with the two central equations of the model. Let yt be the logarithm of output and zt be the logarithm of the potential level of output, the latter being the level of output that would occur in the case of fully flexible wages and prices. The logarithm of the output gap is defined by xt = yt − zt . Furthermore, let πt be the inflation rate, equalling the per cent change in the price level from period t-1 to period t. Let it be the short-term nominal policy interest rate chosen by the central bank, and each period is taken to be a quarter. The intertemporal “IS” curve is the aggregate-demand relation:

3. See Bernanke, Gertler, and Gilchrist (1999) for the derivation of the micro-foundations.

xt = Et xt+1 − 1/σ (it − Etπt+1 − ρtn),

where Et is the expected value operator denoting rational expectations held in period t. The inverse elasticity of intertempo-

261

THE PUBLIC SPHERE | 2017 ISSUE

ral substitution σ corresponds to the interest elasticity of the IS curve. The natural rate of interest ρtn is given by: ρtn = ρ + σgt + σEt (zt+1 − zt),

where ρ > 0 is the constant long-run real equilibrium interest rate, gt is a demand shock, and zt is again the logarithm of potential output. The result is that ρtn is exogenous because both gt and zt are exogenous. According to the IS curve representation of expenditure decisions, higher expected output raises current output because agents smooth consumption over time. Note that it − Etπt+1 is equal to the period t real interest rate and that it − Etπt+1 − ρtn is the deviation of the real interest rate from the economy’s natural rate of interest, sometimes called the interest rate gap. Hence, the output gap is inversely related to positive interest rate gaps. The aggregate-supply relation is the New Keynesian Phillips Curve (NKPC) specified as: πt = Kxt + βEtπt+1 + ut ,

where K is the slope of the NKPC, determined by the level of price stickiness in the economy and satisfying K > 0. A higher value of K implies less stickiness, meaning a higher portion of a given output gap is passed through into inflation. The discount factor β, with 0 < β < 1, indicates how much the representative household discounts the future relative to today. The final term ut is a cost-push shock that embodies exogenous changes in inflation. These include changes in inflation expectations, oil prices, dollar appreciation, and nominal wages that “push real wages

262

[Monetary policy at the zero lower bound ]

away from their ‘equilibrium’ values due to frictions in the wage contracting process” (Evans et al., 2015; Clarida, Galí, and Gertler, 1999, p. 1667). The central bank gauges the desirability of a given policy by how well it minimises the welfare losses associated with deviations of inflation and output from their targets. The loss function for a given time period is: Lt = (πt − π*)² + λ(xt − x*) ²,

where π* is the target inflation rate, x* is the logarithm of the optimal output gap, and the parameter λ > 0 is the weight on output-gap variability relative to inflation variability. These squared deviations of inflation and the output gap from their socially desired levels represent welfare costs to the representative household, which the central bank tries to minimise.4 Without loss of generality, we can assume that the steady-state inflation target and optimal output gap are zero, so that the full intertemporal loss function becomes: ∞ L = 1/2 E₀ Σ βt (πt ² + λxt2),

4. Woodford (2003) derives this loss function from welfaretheoretic foundations as the negative of a quadratic approximation of the utility of the representative household.

t=0

where the discount factor βt is the same as in the NKPC.5 The central bank will prefer the policy that it expects to minimise the discounted welfare losses.

5. However, if we take the intertemporal loss function to represent a central bank’s preferences or its legislative mandate, then the discount factor may diverge from that of the representative household (Woodford, 2014).

263

THE PUBLIC SPHERE | 2017 ISSUE

METHODOLOGY Numerical simulations I perform numerical simulations in a New Keynesian dynamic stochastic general equilibrium model founded in the optimal policy framework discussed in Section 2. It is a forward-looking model assuming rational expectations that I solve for optimal discretionary policy as well as under commitment to variants of the Taylor rule, a method to model a central bank’s public commitment to maintaining a particular monetary policy. The model is solved around a zero-inflation steady state, implying that the ZLB is binding if the natural rate is negative. I extend the model programmed in MATLAB by Evans et al. (2015) in order to include natural interest rate shocks and the calculation of welfare losses with a different loss function. It is a closed-economy model and no fiscal policy shocks are assumed, although they can be incorporated as aggregate demand shocks. Given that unconventional monetary policy is an imperfect substitute for the policy rate, I abstract from large-scale asset purchases and assume that the ZLB is a hard constraint. Since the central bank I study here is discretionary, it cannot do “Odyssean” forward guidance, in which the central bank publicly commits itself to specific future interest rate movements in the future. A discretionary central bank cannot credibly tie its hands to prevent itself later reneging, so financial markets will not believe any such promises about the future. Rather, the central bank engages in “Delphic” forward guidance, in which it publicly states forecasts of the economy and likely policy (Campbell et al., 2012). Effectively, the central bank publishes its forecasts of inflation, GDP growth, and how it would conduct monetary policy if and when such forecasts turn out to be correct. Since this model is forward-looking and solved by backward induction from a known terminal steady state, the assumption

264

[Monetary policy at the zero lower bound ]

of “rational expectations” entails perfect private-sector foresight. Policy paths are anticipated, as are shocks to the economy. These simulations should therefore be interpreted in the spirit of Laséen and Svensson’s (2011) description of anticipated policy paths. They are equilibria that “correspond to situations where the central bank transparently announces that it, conditional on current information, plans to implement a particular policy-rate path and where this announced plan for the policy rate is believed and then anticipated by the private sector” (Laséen and Svensson, 2011, pp. 1-2). For convenience, I repeat the model here in the minimization form: ∞ min 1/2 E₀ Σ βt (πt ²+ λxt2) it

t=0

subject to: xt = Et xt+1 − 1/σ (it − Etπt+1 − ρtn), πt = Kxt + βEtπt+1 + ut , it ≥ 0. Since the central bank is discretionary, and its policy decisions do not affect future losses, it simply minimises πt2 + λxt2 each period. The possibility of a slightly negative policy rate is ruled out by the non-negativity constraint on the policy rate. For the numerical simulations, I follow the solution methods of Evans et al. (2015), both for the discretionary policy and the Taylor rules.6 Their parameters for the slope of the Phillips

6. For the solution methods of Evans et al. (2015), see their Appendix 1 here: https://sites.google. com/site/fgourio/ research_chrono/ publications.

265

THE PUBLIC SPHERE | 2017 ISSUE

curve K, the inverse elasticity of intertemporal substitution σ, and the discount factor β “are all set to values common in the New Keynesian literature” and therefore maintained in the present study (Evans et al., 2015, pp. 159). Appendix A (online) contains the full list of parameter values. I calibrate the model to reflect the state of the economy of the United States, with t = 0 being the first quarter of 2016, and the simulations occur over six years. The assumptions about the exogenous variables are as follows. After the last date, t = 24, there is no uncertainty in the model. Therefore, for t ≥ 24, the cost-push shock ut = 0, and the natural rate ρtn equals the long-run equilibrium real rate ρ = 1.3 per cent. The value of ρ corresponds to the Federal Reserve’s median longrun policy rate from March 2016 minus the Federal Reserve’s inflation target π* = 2 per cent. For t < 24, the natural rate ρtn has a deterministic part and a random part: ρtn = ft + εt

7. The natural rate’s random component εt is an AR(1) stochastic process with a standard deviation of innovation σε=1.32 and a serial correlation of ρε=0.85. The cost-push shock ut is also an AR(1) stochastic process with a standard deviation of innovation σu=0.1 and a serial correlation of ρu=0. The cost-push shock is independent and identically distributed. See Evans et al. (2015, p. 19) for a discussion of this parameterization.

266

The deterministic part ft rises linearly toward ρ = 1.3 per cent. Also for t < 24, the cost-push shock ut is an exogenous mean-zero stochastic process.7 I keep the calibration of Evans et al. (2015) for the initial natural rate of ρ1n = -0.5 per cent, which is similar to the estimates in Laubach and Williams (2015, Figures 5 and 10). From the perspective of period t, the central bank observes both the value of the current natural rate ρtn and cost-push shock ut. Since the starting natural rate is negative, the ZLB initially binds, and the question is when to lift the policy rate from zero. The model is solved by backward induction with 50,000 simulations drawn from the distributions of the stochastic elements.

[Monetary policy at the zero lower bound ]

Policy reaction functions The key addition Evans et al. (2015) make is allowing the central bank to consider uncertainty about the natural rate and, hence, the probability of the policy rate being constrained by the ZLB. This change is a departure from the certainty equivalence assumption in many models, in which the environment is assumed to be deterministic, and optimal policy responds only to the mean of the target variables instead of considering the distribution of future disturbances (Evans et al., 2015; Woodford, 2003). If the natural rate is negative, then the policy rate is set to zero. If the natural rate is positive, however, the optimal path for the policy rate is given by this reaction function: it = ρtn + Etπt+1 + σ (Etxt+1 − xt).

This reaction function is an example of what Svensson (2003) calls an implicit reaction function, as it is a relationship between the policy rate and endogenous, non-predetermined variables. In the context of the full model, this function is an equilibrium condition that must be “solved together with the rest of the model in order to determine the instrument setting” (Svensson, 2003, pp. 14). While the current natural rate ρtn is observed by the central bank, the other variables on the right-hand side depend on unknown expected values of future conditions of the economy and the policy rate. Specifically, uncertainty about the natural rate enters the policy rate decision because xt, Etxt+1, and Etπt+1 all depend on uncertain future values of the natural rate. To account for the risk of a negative natural rate and a binding ZLB from this uncertainty, Evans et al. (2015) model risk management by truncating the distribution of the natural rate when they calculate the expected values of endogenous variables.

267

THE PUBLIC SPHERE | 2017 ISSUE

Hence, the endogenous variables are only affected by the negative part of the distribution of the natural rate. I maintain this treatment of the asymmetric risk because a positive natural rate in future periods can be offset simply by traditional increases in the policy rate. A small but instructive addition to this optimal discretionary policy is to assume that the policymaker is a “Rogoff conservative” central banker. In this case, the central bank puts significantly higher weight on stabilizing the inflation gap rather than the output gap, and the private sector understands this (Rogoff, 1985). While this method was originally developed to tackle the positive cost-push shocks of the 1970s that created high inflation, it is also useful in a liquidity trap. Essentially, restoring inflation from below target is very important for lowering long-term real interest rates and spurring consumption today. I implement this by decreasing the relative weight λ on the output gap in the loss function that is used for deriving the optimality conditions from λ = 0.25 to a lower Rogoff-style value of λR = 0.1. Even though λR is less than the “true” λ in society’s loss function, this change can yield better overall outcomes by effectively changing how the private sector expects the central bank to respond to disturbances. I follow Evans et al. (2015) in comparing the optimal discretionary policy with three other reaction functions. The first is a “naïve” policy that is the same as optimal discretion, except the central bank does not consider uncertainty about the future. The private sector knows this but still considers uncertainty itself. The central bank is then surprised in two ways: the arrival of each new shock and the realised values of πt and xt, which the private sector determined with rational expectations. The two reaction functions representing the commitment case are both forms of the Taylor rule, a formula the central bank publicly commits to use for setting the policy rate systematically. Such a scenario is instructive for comparison with the discretionary case because it models the economy’s evolution under the ex-

268

[Monetary policy at the zero lower bound ]

treme assumption of fully credible commitment. The first Taylor rule is the original Taylor (1993) formulation with the constant long-run equilibrium real interest rate ρ, and the second reacts to the short-term natural rate ρtn: Taylor rule with constant LR rate (TRCI): it = ρ + π* + ϕ πt + yxt

Taylor rule with time-varying natural rate (TRTV): it = ρtn + π* + ϕ πt + yxt.

The weights on the inflation gap ϕ and the output gap y are set to 1.5 and 0.5 respectively, as in Taylor (1993). The other notation remains the same as above. Apart from the constant long-run equilibrium real interest rate ρ and constant inflation target π*, both the TRCI and TRTV are set by the central bank immediately in response to that period’s predetermined variables. The weights imply that the policy rate rises by 1.5 times the inflation gap and 0.5 times the output gap. The TRTV’s one-to-one response with the short-term natural rate ρtn means that it responds more accurately to the level of activity in the economy, as opposed to the TRCI’s use of the unchanging long-run equilibrium real interest rate ρ. To assess concerns of welfare losses from policy rate volatility, I run the same simulations again with a standard addition to the loss function to penalise changes in the policy rate (Woodford, 2003). This addition is useful for determining whether an otherwise- desirable policy only stabilises inflation and output through

269

THE PUBLIC SPHERE | 2017 ISSUE

intolerably high variation in the policy rate. The per-period loss function becomes: Lt = (πt − π*)² + λ(xt − x*) ² + λ∆i(it − it-1).

The parameter λ∆i is the weight placed on changes in the policy rate between periods (it − it-1). This extra term in the loss function indicates that the central bank has a preference for interest rate smoothing. The weight λ∆i is usually set to 0.1 or 0.5 in the literature (Rudebusch, 2006; Alichi et al., 2015). Although aversion to changes in the policy rate does not enter society’s true loss function, adding this term into the central bank’s loss function can be motivated by a desire to stabilise the real economy as well as observed smoothing by central banks. For instance, Brainard (1967) demonstrates that gradual changes in the policy rate are beneficial when there is uncertainty about how the policy will affect the economy, and high volatility in bond markets could hinder financial intermediation and the credit-supply process (Alichi et al., 2015). Woodford (2003) also argues that more gradual monetary policy increases its predictability, allowing small changes in the policy rate to cause large changes in long-term interest rates. The problem then follows that central bank gradualism may not prevent volatility in longer-term interest rates because a given change in the policy rate is amplified when interpreted by the markets as a signal of future changes to come (Stein and Sunderam, 2015). Rudebusch (2006) gives two arguments for smaller weights on policy rate changes. First, the United States had high interest rate volatility during the 1979-82 monetary experiment, and this did not generate considerable costs on its own. Second, the weights of λ∆i = 0.1 or 0.5 “still seem at the high end of the plausible range of penalties to reduce volatility, especially in a world

270

[Monetary policy at the zero lower bound ]

with a wide variety of financial market instruments that allow for hedging against interest rate volatility” (Rudebusch, 2006, p. 107). With this disclaimer in mind, I report estimates with both weights.

SIMULATION RESULTS Figure 1 displays the modal outcomes for the evolution of the economy in the baseline re-calibrated model starting in the first quarter of 2016.8 The black horizontal lines are the steady-state values of each variable. For the policy rate, the horizontal line is the long-run policy rate projected by the Federal Reserve, which is equal to the long-run equilibrium real rate ρ = 1.3 per cent plus the inflation target of 2 per cent. Note that the plot in which the optimal discretionary policy uses the Rogoff conservative weight in the loss function is reported in Appendix B (online). There are no shocks in the baseline simulation. The optimal discretionary policy “lifts off ” from the ZLB later than the other policies. The naïve policy simply tracks the natural rate’s increase throughout the simulation, so the difference between the green and red interest rate paths illustrates how uncertainty about the ZLB binding impacts discretionary policy. This difference is between 0.5 and 1.6 percentage points over the first two years. The inflation outcomes for the naïve policy are much worse because it maintains a high policy rate early, but the output gap outcomes under the optimal discretionary and naïve policies are very similar. Of the two Taylor rules, the one with the time-varying natural rate ρtn (TRTV) performs much better than the original formulation (TRCI) of Taylor (1993), which includes the constant long-run equilibrium real rate ρ. Since ρ is by definition larger than ρtn here, and the rules are otherwise identical, it is clear that the TRTV will prescribe a more gradual increase. The poor inflation and output gap performance of the TRCI is due to

8. In all graphs here, I follow Evans et al. (2015) in reporting the modal outcome; this measure is useful because the Federal Open Market Committee members report their modal forecasts of monetary policy in their Summary of Economic Projections.

271

THE PUBLIC SPHERE | 2017 ISSUE

the fact that the private sector anticipates the higher future policy rate, and this dampens consumption and inflation today. FIGURE 1 EVOLUTION OF KEY VARIABLES 5

Nominal interest rate

1.5

% 1 0.5

% 2 1

0 -1

Inflation

2.5

-0.5 5

15 Quarters

-1

15 Quarters

Output gap

0.5 0

Optimal Discretion Naïve Taylor Rule Time Varying Taylor Rule Constant

-0.5 % -1 -1.5 -2

10 15 Quarters

Source: Author.

Table 1 Simulation results Statistic

Naïve

Taylor Rule (TRCI)

Taylor Rule (TRTV)

Loss

0.03

0.12

0.14

0.04

Loss + 0.1 (it − it-1)

0.10

0.14

0.18

0.17

0.10

Loss + 0.5 (it − it-1)

0.39

0.56

0.41

0.29

0.34

Mean time to liftoff

4.63

5.16

Median π at liftoff

1.72

1.65

0.21

0.64

1.34

Median x at liftoff

0.05

0.02

-2.19

-1.33

-0.43

2.64

2.55

2.27

2.43

2.05

-0.91

-1.04

-2.19

-2.51

-1.10

1.89

3.06

1.22

1.04

1.39

Median time to liftoff

75th percentile of max π 25th percentile of min x Median std. dev. of �i

272

Optimal Opt. Disc. Discretion Rogoff

[Monetary policy at the zero lower bound ]

With “liftoff ” defined as the policy rate being increased above 0.25 per cent, all policies other than optimal discretion and its modified Rogoff formulation leave the ZLB in the first quarter. Optimal discretion does not raise the policy rate until a median of the 2 quarters or a mean of 3.63 quarters later than the other policies. In the Evans et al. (2015) study, the analogous delay at the ZLB under optimal discretion is two to three quarters after the other policies rise. The extra delay in this simulation is due to the fact that the long-run projected policy rate has been adjusted downward from 3.5 to 3.3 per cent, reflecting weaker aggregate demand than expected. With the modification for the Rogoff conservative central banker, optimal discretion performs largely similarly, with two exceptions. Firstly, it’s mean time for raising the policy rate is even later, reflecting the central bank’s desire to restore inflation to target quickly. Secondly, it has higher volatility in interest rate changes. The welfare losses in Table 1 show that optimal discretion performs very well according to the original loss function, incurring 1/4 the loss of the naïve policy, 1/5 the loss of the TRCI, and 3/4 the loss of the TRTV. With a weight of λ∆i = 0.1 on interest rate variation, the extra losses imposed by optimal discretion put it on par with the TRTV, while the other policies still perform much worse. However, the ranking changes with the higher weight of λ∆i = 0.5. The TRCI minimises losses because it is the least volatile—with a median standard deviation of the change in the interest rate of 1.04. For reference, Evans et al. (2015) empirically observe this standard deviation of changes in the Federal Reserve’s policy rate to be 0.88. Two caveats pertain to this analysis of losses from policy rate variation. First, the results are clearly sensitive to the assumed weight λ∆i. Second, policies with a higher initial policy rate have the advantage of needing smaller policy rate changes to reach the long-run nominal policy rate of 3.3 per cent. In the first quarter of 2016, the Federal Reserve’s policy rate was at a corridor of

273

THE PUBLIC SPHERE | 2017 ISSUE

0.25-0.50 per cent, so a policy with a greater starting value than 0.50 per cent underestimates the losses here. Lastly, the policies balance output and inflation risks differently. Across simulations the 75th percentile of maximum inflation is only slightly higher under optimal discretion than under the other policies, and the 25th percentile of the minimum output gap under optimal discretion is slightly less than the TRTV and less than half that of the other policies.

CONCLUSIONS The central theoretical contribution of this paper is that an optimal discretionary policy accounting for uncertainty about the natural rate forcing the ZLB to bind can perform as well or better than commitment to a Taylor rule, even though the discretionary policy has a more volatile policy rate. The validity of this finding depends on the true costs to society from changes in the policy rate and our ability to hedge against them. If these costs are high relative to the benefits of the “lower for longer” policy in resolving the liquidity trap, then a more gradual policy that begins raising rates earlier is desirable. An addendum to this finding is that the appointment of a Rogoff conservative central banker may be beneficial for discretionary central banks at or near the ZLB. While the Rogoff formulation is associated with keeping inflation expectations from running too high, a central bank that is symmetric about its inflation target will also diminish the deflationary bias of discretion at the ZLB. This last condition should not be taken for granted, as lately some central banks appear willing to tolerate negative inflation gaps more than positive ones of equal magnitude. Nonetheless, the cost from Rogoff ’s formulation is variation in the output gap and policy rate. It is important to note limitations of this study before applying the findings to the real world. The model does not include

274

[Monetary policy at the zero lower bound ]

unconventional policy in the form of large-scale asset purchases, â&#x20AC;&#x153;Odysseanâ&#x20AC;? forward guidance, or negative policy rates. These omissions overstate the severity of the ZLB and portray it as a harder constraint than it is in reality. If the model were changed to include them, the effective lower bound would be slightly negative, liftoff would be earlier, and the deflationary effects of the liquidity trap would be dampened. Fiscal policy expansion is another tool that can support aggregate demand and help end a liquidity trap (Krugman 1998). I omit fiscal policy from the analysis both for parsimony and the fact that fiscal policy is subject to the political process, which means it is not a readily available tool for economic policymakers. The model has no international sector, which leaves out another important part of the policy decision. Inclusion of an international sector representing the economic context of downward growth forecasts in early 2016 heightens the risks of the ZLB. Lastly, the assumption of perfect foresight stemming from rational expectations likely does not accurately describe times of such unprecedented monetary policy, as expectations cannot now be informed by past experience. As I have calibrated the model to conditions in the United States, I discuss the policy implications of this study in the context of the Federal Reserve. The implications, however, are instructive for the various central banks with policy rates still at the ZLB. The Federal Reserve is a worthy case to study not only due to the dominance of the dollar but also due to its December 2015 decision to raise the policy rate from the zero lower bound. In September 2015, the Federal Reserve did not raise rates despite previous forward guidance that had seriously opened the possibility, and the Federal Reserveâ&#x20AC;&#x2122;s subsequent rhetoric may have boxed it into the December 2015 decision to raise the policy rate corridor from 0.0-0.25 per cent to 0.25-0.50 per cent (Feroli et al., 2016). The 2015 median personal consumption expenditures inflation of 1.3 per cent, also published during the same December 2015 meeting, was substantially below the central

275

THE PUBLIC SPHERE | 2017 ISSUE

bank’s 2 per cent inflation target. Despite its failure to achieve the 2 per cent inflation target, the Federal Reserve has made much better progress on its employment objective since the financial crisis. The unemployment rate was 4.9 per cent during the first two months of 2016. There seems to be some room for improvement here too, however, as the labour force participation rate is still not back to its pre-crisis level (Appendix C, online). From the perspective of this study’s numerical simulations, the Federal Reserve raised rates slightly early. One would expect this given that my calibration uses their updated March 2016 projection of the long-run equilibrium real interest rate that was revised downward from the December 2015 estimate. Of course, the Federal Reserve cannot know the future; the slowdown in China, and emerging markets in general, as well as further declines in the oil price contributed to the downward revision of U.S. growth and inflation projections from December 2015 to March 2016 (FOMC, 2016). In light of these developments, the Federal Reserve has not raised rates again yet (through August 2016). Chair Janet Yellen’s speech at the end of March 2016 includes separate sections for risks to the real economy and the inflation outlook. Indeed, with a title of “The Outlook, Uncertainty, and Monetary Policy” the speech indicates a movement toward risk management in the spirit of Evans et al. (2015), as opposed to the more calendar-based forward guidance employed since the crisis (Feroli et al., 2016). In sum, this study implies a need for heightened risk management near the ZLB and a lower policy rate until the recovery advances further and the risk asymmetry moderates. Concerns about the volatility of a policy that remains at the ZLB longer are not supported by the model, nor are concerns that such a policy would be unable to respond to a burst of inflation. Lastly, the public perception of a Rogoff conservative central banker who cares equally about stabilizing negative and positive inflation gaps helps lift inflation expectations toward target.

276

[Monetary policy at the zero lower bound ]

REFERENCES Adam, K., and Billi, R.M., (2005). Discretionary monetary policy and the zero lower bound on nominal interest rates. (Research Working Paper RWP 05-08). Kansas City: Federal Reserve Bank of Kansas City. Alichi, A., Clinton, K., Freedman, C., Kamenik, O., Juillard, M., Laxton, D., Turunen, J., Wang, H. (2015). Avoiding Dark Corners: A Robust Monetary Policy Framework for the United States. (IMF WP/15/134). Washington D.C.: International Monetary Fund. Bank of Canada (2015). Bank of Canada updates framework for unconventional monetary policy measures. Policy statement, 8 December 2015. Bernanke, B. (2012). Monetary Policy since the Onset of the Crisis. Speech at the Federal Reserve Bank of Kansas City Economic Symposium, Jackson Hole, Wyoming. Bernanke, B.S.; Gertler, M., and Gilchrist, S. (1999). The Financial Accelerator in a Quantitative Business Cycle Framework. In: Taylor, J. B. and Woodford, M. (Ed.). Handbook of Macroeconomics, (pp. 1341-1393). Elsevier. Brainard, W. (1967). Uncertainty and the Effectiveness of Policy, American Economic Review. 57, 411-425. Calvo, G.A. (1983). Staggered Prices in a Utility Maximizing Framework. Journal of Monetary Economics. 12 (1983) 383-398. Campbell, J.R., Evans, C.L., Fisher, J.D.M., Justiniano, A. (2012). Macroeconomic Effects of Federal Reserve Forward Guidance. Brookings Papers on Economic Activity, Spring 2012. Clarida, R., Galí, J., and Gertler, M. (1999). The Science of Monetary Policy: A New Keynesian Perspective. Journal of Economic Literature, 37(Dec. 1999), 1661–1707. Eggertsson, G. (2006). The Deflation Bias and Committing to Being Irresponsible. Journal of Money, Credit, and Banking, 38 (2), 283-321. Evans, E., Gourio, F., Fisher, J., and Krane, S. (2015). Risk Management for Monetary Policy Near the Zero Lower Bound. Brookings Papers on Economic Activity. 50 (1), 141- 219. Feroli, M., Greenlaw, D., Hooper, P., Mishkin, F.S., Sufi., A. (2016). Language after Liftoff: Fed Communication Away from the Zero Lower Bound. Prepared for the 2016 US Monetary Policy Forum, February 26. FOMC (2016). Minutes of the Federal Open Market Committee, January 26–27, 2016. Washington, D.C.: Board of Governors of the Federal Reserve System Krugman, P. R., (1998). It’s Baaack: Japan’s Slump and the Return of the Liquidity Trap. Brookings Papers on Economic Activity, 29 (2), 137–206. Laséen, S., and Svensson, L.E.O. (2011). Anticipated Alternative Policy-Rate Paths in Policy Simulations. Riksbank Research Paper Series No. 79. Stockholm: Sveriges Riksbank. Laubach, T., Williams, J. (2015). Measuring the Natural Rate of Interest Redux. Federal Reserve Bank of San Francisco Working Paper 2015-16. Orphanides, A. (2003). Monetary Policy in Deflation: The Liquidity Trap in History and Practice. Finance and Economics Discussion Series, 2004-01. Washington D.C.: Board of Governors of the Federal Reserve System. Rogoff, K. (1985). The Optimal Degree of Commitment to an Intermediate Monetary Target. Quarterly Journal of Economics, 100 (Nov.), 1169–1190. Rotemberg, J.J., and Woodford. M. (1997). An Optimization-Based Econometric Framework for the Evaluation of Monetary Policy. NBER Macroeconomics Annual 1997, 297–346.

277

THE PUBLIC SPHERE | 2017 ISSUE

Rudebusch, G.D. (2006). Monetary Policy Inertia: Fact of Fiction? International Journal of Central Banking, 1 (4), 85-135. Schmitt-Grohe, S., and Uribe, M. (2004). Optimal Operational Monetary Policy in the Christiano-Eichenbaum-Evans Model of the U.S. Business Cycle. NBER Working Paper 10724. Stein, J.C., and Sunderam, A. (2015). Gradualism in Monetary Policy: A Time-Consistency Problem? NBER Working Paper 21569. Svensson, L.E.O. (2003). What is Wrong with Taylor Rules? Using Judgment in Monetary Policy through Targeting Rules. Journal of Economic Literature, 41(2): 426-277. Svensson, L.E.O. (2006). Monetary Policy and Japan’s Liquidity Trap. (CEPS Working Paper No. 126). Prepared for the ESRI International Conference on Policy Options for Sustainable Economic Growth in Japan, Tokyo. Taylor, J.B., (1993). Discretion versus Policy Rules in Practice. Carnegie-Rochester Conference Series on Public Policy, 39 (Dec.), 195-214. Woodford, M. (2003). Interest and Prices: Foundations of a Theory of Monetary Policy. Princeton: Princeton University Press. Woodford, M. (2010). Optimal Monetary Stabilization Policy. In Friedman, B.J., and Woodford, M. (Ed.). Handbook of Monetary Economics. 723-828. Elsevier. Yellen, J. (2014). Many Targets, Many Instruments: Where Do We Stand? In Akerlof, G., Blanchard, O., Romer, D., and Stiglitz, J. (Eds.). What Have We Learned? Macroeconomic Policy after the Crisis. MIT Press. Yellen, Janet L. (2016). The Outlook, Uncertainty, and Monetary Policy, the Economic Club of New York, New York, March 29. The Chamber of Commerce of the United States. (2006). Land Transport Options Between Europe and Asia: Commercial Feasibility Study. Retrieved 3rd May 2015, from http://www. osce.org/eea/41310?download=true. Winterbottom, V. (2012). From Camels to Horsepower. Beijing Review. 17th September. Available from: http://www.bjreview.com.cn/Cover_Stories_Series_2014/2012-09/17/ content_609212_2.htm. [Accessed: 3rd May 2015].Winterbottom, V. (2012). The Silk Railroad. China Today. Retrieved 3rd May 2015, from http://eds.b.ebscohost.com/ ehost/pdfviewer/pdfviewer?sid=78a8641b-f4af-4d4a-a99f-7f895bb7df87%40sessionmgr111&vid=5&hid=112.

Please see publicspherejournal.com for online appendices.

278

Flushing, rituals and needle fixation among heroin addicts Implications for policy John Wooldridge

MSc Social Policy (Research), Class of 2016 London School of Economics and Political Science

ABSTRACT Research has found that needle fixation is a genuine problem and has an impact on achieving abstinence for a minority of heroin users. For those with these dual addictions, it is even more difficult to stop using. Previous studies have focused on defining the problem of needle fixation and attempting to measure its prevalence. This study seeks to uncover the reasons why some heroin users develop needle fixation and how needle fixation, flushing, and rituals develop over time. It also seeks to make policy suggestions on how to improve heroin addiction treatment. To answer these questions, indepth interviews are conducted with heroin addicts in the UK, and the results are thematically analysed. It is discovered that people with needle fixation have other obsessive traits, which contribute to the development of the phenomenon. It is found that rituals develop incrementally over time, because of advice given by friends, as well as contradictory advice from professionals. Suggested policy interventions include reintroducing injectable methadone and increasing the number of available residential rehabilitation spaces, as these are rarely used in the UK. One service should be in charge of addiction treatment to ensure clarity in terms of advice and accountability. A suggested avenue for further research would be to conduct a quantitative study and attempt to assess how prevalent needle fixation is among heroin addicts.

[Flushing, rituals and needle fixation among heroin addicts]

INTRODUCTION Needle fixation is defined as the “repetitive puncturing of the skin with or without the injection of psychoactive drugs via intravenous, subcutaneous or intramuscular routes, irrespective of the drug or drugs injected or the anticipated effects of the drugs” (Pates et al. 2001, 15). Drug users with symptoms of needle fixation have been seen all over the world, but mainly in the UK, Europe, and the USA (Pates et al. 2005). This research is specifically focused on heroin users with needle fixation because heroin is the most common reason for seeking addiction treatment in the UK. Among those seeking treatment in the UK in the last year for drug abuse, heroin addiction was cited in 47 per cent of cases, while 32 per cent of cases involved the use of both heroin and crack (PHE 2014). Heroin is the most common injectable drug and the drug most commonly linked to needle fixation in the public health literature. There is evidence to suggest that needle fixation is also linked to ritualistic practices and needle flushing, each of which pose additional complications for treatment and health. Previous research has found that some heroin users with needle fixation persistently ‘flush’ their syringe (pulling and pushing blood through the syringe after the drug had been administered) or have rituals in the preparation of their drugs (Pates and Gray 2009; Stewart 1987; Hinton et al. 2013). The aim of the present study is to explore the concept of needle fixation, specifically among people that use heroin. Needle fixation is an academically recognised phenomenon, however, little acknowledgement has been seen in terms of treatment or adopted in rehabilitation programmes (Treffurth and Raj Pal 2010). Because heroin use and habitual injection represent mutually reinforcing addictions, the interruption of the combined behaviors becomes inherently more difficult.

281

THE PUBLIC SPHERE | 2017 ISSUE

The existing research finds that needle fixation could prevent some people from abstaining from injected drug use, however the reasons why only some people develop these fixations are unknown (McBride et al. 2001). Through in-depth interviews with heroin users, this paper discusses why some individuals develop needle fixation. Furthermore, this study determines how these processes develop and whether needle fixation is a problem that needs to be directly addressed within treatment programs. Through an analysis of these qualitative interviews, the research provides specific policy recommendations to improve the care mechanisms used to treat addicts with needle fixation in the UK.

LITERATURE REVIEW Development of UK drug policy In the 1980s, due to the significant increase in the use of heroin and the beginning of the AIDS epidemic, “harm reduction” gained prominence as a focus of drug policy both in the UK and internationally. This led to many new initiatives aimed at decreasing the spread of the disease (Advisory Council on the Misuse of Drugs 1988). Methadone maintenance treatment (MMT) was provided as a daily opioid substitute, usually taken orally to offer addicted users a legally prescribed alternative to street heroin (Gossop 2003). The aim was to stabilise users and reduce their use of illegal drugs without triggering withdrawal symptoms (Ward et al. 1999). Initially, methadone was prescribed by general practitioners (GPs). However, following the 1982 Advisory Council on the Misuse of Drugs (ACMD) report, Treatment and Rehabilitation, the government created a funding initiative to implement a non-statutory drugs service for people in most areas. This

282

[Flushing, rituals and needle fixation among heroin addicts]

service has been adopted across the country, but services differ from place to place. Local authorities outsource their drug treatment obligations by putting contracts out to a tendering process, so services are delivered by non-government bodies. This means that contracts are often given to the cheapest provider, which may be detrimental to the service user. Most people on MMT now receive their treatment through a specialised addictions service, but people with health problems can still be treated by their GP. Since 2012, the government has opted for a more abstinence-based approach to treatment – abstinence meaning the complete removal of all opiates from the user (HM Government 2010; Home Office 2012). Abstinence-based treatments are far more costly than MMT as these involve the user being moved to a residential rehab centre for a complete physical detox. Annually, abstinence-based treatment methods cost approximately £8,000 per user, while prescribing a user methadone incurs a cost of £3,000-£5,000 (Laurence 2009). However, the results from abstinence-based methods are far more favourable, as they have a success rate of 28 per cent compared to 4 per cent for MMT (National Treatment Agency, 2012). Recent increases in support for abstinence-based treatment have been fuelled in part by the argument that MMT is a way to control people, rather than help them become drug-free (Lally 2013). Another issue concerning methadone prescription is known as the “parking” problem: over the years, methadone has been over-prescribed, and giving people a daily methadone prescription does not help people “sort out other issues in their lives” (DrugScope 2009, 20). The National Treatment Agency (NTA) said that the number of people “parked” on methadone for four months or more increased by one quarter between 2010–2011 and 2012–2013, from 39,725 to 48,510 (PHE 2013). The most recent statistics show that just 7 per cent of enrolled people completed opiate addiction treatment and did not return

283

THE PUBLIC SPHERE | 2017 ISSUE

to the service within six months in 2014–2015 (National Drug Treatment Monitoring System 2016). Notably, in that same period, four per cent of people died while receiving treatment; the same proportion as those who completed treatment free of prescription drugs (NDTMS 2015).

Needle fixation Estimates suggest that needle fixation is observed among a quarter of injecting heroin users (Hinton et al. 2013). Pates et al. (2001) conducted a literature review on needle fixation citing the extensive library on drugs and addiction at the Institute for the Study of Drug Dependence in London. Pates et al. concluded that needle fixation should be taken seriously due to concerns about blood-borne viruses among intravenous drug users and a critical need to reduce the practice of injecting. Among heroin addicts, the habitual injection of drugs makes needle fixation a conditioned response whereby users achieve a secondary form of stimulation. Ultimately, this secondary form of stimulation, triggered by the preparation of the injection and the injection itself, helps to maintain the practice of drug use (Pates et al. 2005). Stimulating practices can include the routine of acquiring money to buy drugs, the ritual of arranging and administering the drug, and injecting with substitutes such as water, in the absence of drugs. Pates et al. (2001) recommend that a greater understanding of the problems posed by fixation will help in the effective treatment of injecting drug users. In his 2002 study, Hampl further corroborates these findings, noting that some opiate addicts, adequately dosed with methadone, still felt cravings to inject. Like Pates, he identifies fixation as a conditioned state due to the ritualistic nature of heroin injecting and thus argues that existing drug dependence treatment programs do not sufficiently acknowledge or adapt to the role of ritual in addiction.

284

[Flushing, rituals and needle fixation among heroin addicts]

Drug users also describe needle fixation as a barrier to change. A qualitative study by McBride et al. (2001) found that some users had a compulsive need to inject despite knowing there would be no psychoactive effects. For those who exhibit a seemingly compulsive need to inject, interventions need to be “tailored to address their individual reasons for doing so” (McBride et al., 2001, p.1057). Thus, failure to recognise and address the complex determinants that reinforce obsessive behaviour among some users may limit the effectiveness of drug treatment programs. Pates et al. (2009) furthered this work by conducting qualitative interviews with drug users in Cardiff. They used this research to create NEFPRO, a framework for describing needle fixation “profiles” of users for applied clinical use. NEFPRO was developed into a 10-point scale of addiction, which can identify needle fixation among drug users and specifies the role of needle fixation for an individual’s drug use pattern based upon their completion of a questionnaire. The researchers conclude that this scale may help with the treatment of injecting drug users by identifying needle fixation earlier in treatment. In an extension of this research, Hinton (2015) conducted an exploratory study to examine the applicability of this needle fixation measure in Australian drug treatment practice. Using the NEFPRO questionnaire, the study assessed whether users were familiar with the pleasurable secondary effects of the injection itself, in addition to the effects of the drug that have been found to be associated with needle fixation. It found that Australian drug users were aware of, and had experience of themes such as flushing, substitution, and injecting water. The authors recommended the use of the NEFPRO questionnaire to other practices in Australia as a measure of needle fixation to ensure that all avenues of harm reduction be explored.

285

THE PUBLIC SPHERE â&#x20AC;&#x2020;| â&#x20AC;&#x2020;2017 ISSUE

Attachments Although not explicitly linked to needle fixation, field research suggests a link between insecure attachments and drug use. Attachment theory is a psychological premise that describes the dynamics of relationships between parents and their children. John Bowlby (1969) stated that children with secure attachments show distress when a caregiver leaves but are able to quickly compose themselves, knowing that the caregiver will return. Children with insecure attachments also show distress when their caregiver leaves but are not able to compose themselves. Additionally, children that have secure attachments to their caregiver will have higher self-esteem and lower reports of anxiety and depression (Hogg and Vaughan 2008). There is evidence to suggest a link between such insecure attachments in childhood and adult substance abuse. In a qualitative study of 19 participants, Borhani (2013) found that there is a positive correlation between substance abuse and types of insecure attachment. Kassel et al. (2006), meanwhile, found that individuals with insecure attachments may lack the necessary skills to form social relationships, which may cause distress or anxiety, which may in turn lead to drug use. In a 2006 study, Thorberg and Lyvers found that, out of a sample of alcohol, heroin, and cannabis addicts, users reported higher levels of insecure attachments than observed in non-users. In a sample of adoptees, Caspers et al. (2005) found higher levels of illicit substance use among insecure attachment groups compared to those with secure attachments. This is interesting as many drug addicts come from unstable family backgrounds, which may explain a driver of their addiction or gravitation towards use (Khoury et al. 2010).

286

[Flushing, rituals and needle fixation among heroin addicts]

METHODOLOGY Data collection strategies Data was collected through in-depth qualitative face-to-face interviews with current heroin users. Participants were recruited via their attendance at a drug treatment centre in a city in Northern England â&#x20AC;&#x201C; a registered charity responsible for drug treatment in the local area. The city has a population of approximately 200,000 and a lower than estimated number of heroin and crack users per 1,000 compared with the rest of England. The proportion of injecting drug users is higher than the UK national average. Prior to taking part in the interview, participants were given an information sheet explaining the research and an informed consent form to read and sign. As heroin users are usually poorly educated (Puigdollers et al. 2004; Luck et al. 2004), the information sheet and consent form were read out to all interviewees. Interviews were semi-structured, using a topic guide of questions which had to be covered during the conversation. Participants were asked about their drug history, injecting history, a step-by-step description of their injecting process, awareness of needle fixation, and questions about treatment. Conducting interviews with drug users raises a number of ethical concerns (Dixon-Swift et al. 2008). Firstly, the sample represents a vulnerable group. Drug users are more likely to suffer from mental health problems than non-drug users. Furthermore, drug users are also a potentially volatile group and are more likely to end up in prison than non-users due to their drug addictions (UKDPC 2008). Therefore, to help account for these concerns, interviews took place in the drug treatment centre where the participants attended and where they were recruited. To guarantee the safety of the interviewer and the

287

THE PUBLIC SPHERE | 2017 ISSUE

provision of emotional support to the research participant, interviews were carried out in a private room with a member of staff present.

Data analysis strategies

1. The codebook is available from the author upon request.

To analyse the interview data, I use the thematic coding method. Firstly, I transcribed audio recordings, then thoroughly read and re-read in order to build familiarity with the content. The research followed Miles and Hubermans’ (1994) sequential list of analytic moves, progressing through a series of steps including initial coding and writing memos. Data was analysed manually because this was a small study with five participants and the data collected was manageable at this scale (Bryman 2012).1

FINDINGS A number of themes alluded to throughout the interview process are highlighted by quotes from the interviews. Figures 1, 2, and 3 describe the themes and sub-themes identified for each question. These thematic similarities are further addressed in the discussion section that follows.

288

[Flushing, rituals and needle fixation among heroin addicts]

FIGURE 1

Other/ previous compulsive traits Why do some heroin users develop addiction to the injection process?

Irrational superstitions

Insecure attachments

Source: Author. FIGURE 2

Incremental development

Ever-increasing rituals How do some people develop these practices (rituals, flushing and needle fixation)?

Belief that flushing improves the hit Contradictory advice from professionals

Advice from friends

Source: Author.

289

THE PUBLIC SPHERE | 2017 ISSUE

FIGURE 3 Stigma and inconvenience of consuming on site at a chemist Dislike of methadone In what ways could treatment be improved to ensure that those with needle fixation are effectively treated for their addiction?

Injectable methadone

Damage to teeth

Perceive the process as controlling

Residential rehabilitation

Source: Author.

Why do some heroin users develop addiction to the injecting process? The following themes were discovered in response to the question of why some people develop these routines.

Other/ previous compulsive traits All of the respondents described having other obsessive or compulsive behaviours, often related to cleaning. RESP1: “I’m really bad for cleaning, everyone says I’m obsessed”. RESP5: “I always make my bed in the same way.”

290

[Flushing, rituals and needle fixation among heroin addicts]

RESP3: “Yeah cleaning and tidying. I’ve got a very tidy house. You wouldn’t believe the things I do.” RESP4: “I got annoyed with the postman when all the post just goes on the floor… Sometimes I realise how stupid it seems.” One participant went further and stated that they had obsessive-compulsive disorder. RESP2: “I’ve got OCD me. It’s been diagnosed as well… It gets into everything, every aspect of my life is ordered in a specific way.”

Irrational superstitions Some interviewees also spoke about irrational superstitions that affect their daily life. RESP1: “I always avoid walking over three gates on the path ‘cause it’s bad luck”. RESP1: “Touching wood, you know when you want or don’t want something to happen.” RESP5: “I always sleep with my head on the opposite side to the door but there is so many more I do as well.”

Insecure attachments Unprompted, three of the respondents spoke about having insecure attachments when they were children. RESP1: “I don’t speak to my parents. I never have done really, I moved out when I was 13.”

291

THE PUBLIC SPHERE | 2017 ISSUE

When talking about when they started using drugs one person said, RESP3: “When I was younger I just lived with my brother and his mates introduced me to it.” One participant mentioned that they have been in foster care as a child. RESP5: “I spent all my life in care so I was at different care homes really. There was never any stability.”

How do some people develop these practices (rituals, flushing and needle fixation)? All respondents were asked to describe how they inject their drugs step-by-step to try to ascertain every detail of what they did and the reasons for each step. The following themes describe the reasons why certain rituals developed in this group.

Incremental development All respondents described how their injecting practices and rituals had developed incrementally. RESP1: “I never used to do that until recently. Now I do it all the time” RESP2: “My routine was really quick years ago but over time you start doing different things.” RESP4: “When I first started I didn’t have a routine like now.” When talking about developing a long ritual,

292

[Flushing, rituals and needle fixation among heroin addicts]

RESP5: “It just happened. I haven’t really thought about it but it used to be real quick. I don’t know how it got this long.”

Ever-increasing rituals Participants described how their injecting rituals are always increasing. RESP1: “I’m definitely adding more things to it [the injecting routine].” RESP2: “I used to just do it [the injection] but now I run it under the tap for a while too to cool it down.” RESP4: “It takes a long time to do it, ‘cause I like to do so many things. But that’s because I need to make sure I’m doing it right.”

The interviewer asked if there was anything that they no longer do as a part of their routine. RESP1: “There’s nothing I don’t do now that I used to.” RESP4: “It is a bit annoying. It’s like, if you find something that works, you don’t stop doing it.” RESP3: “The only thing I don’t do now is lick the pin ‘cause of the germs.”

Belief that flushing improves the hit Most interviewees stated that the reason they flushed was because it improved the hit and meant the drugs had a greater effect.

293

THE PUBLIC SPHERE | 2017 ISSUE

RESP2: “It gives you a better knock if you flush it.” RESP1: “It definitely makes a difference. I feel way more smashed if I flush it a few times.” RESP3: “Of course it does [improve the effect], your making sure everything is out of the pin.” RESP4: “I do like the feeling but I do it because it gets me more wiped out.”

Contradictory advice by professionals Contradictory advice by professionals was given as a reason for a number of practices. This issue arises where messages from service providers are inconsistent, either over time or between individuals; these inconsistencies are rarely explained or justified, which is disempowering in addition to being confusing for users. RESP1: “So I started using the wipes before I did it [the injection] but then they stopped giving them out and said it’s not advised. Apparently the alcohol isn’t good for wounds. But now I’m still using the wipes ‘cause I got used to it.” RESP3: “They say boil a kettle and use that water, but what do you put it in? It doesn’t make sense… If everything in the house is dirty, should I boil a kettle and put it in a dirty cup?” RESP4: “They used to give out water vials but they stopped ‘cause it cost them too much money. I think 50p each or something. It’s only homeless that get them now.”

294

[Flushing, rituals and needle fixation among heroin addicts]

Despite being harmful, two participants spoke about behaviours that would not be professionally advised. RESP5: “I always lick the end of the pin before I do it”.

Advice from friends Most participants increased and changed their injecting routine after taking advice from friends. RESP2: “If you change the filter you can share a spoon.” RESP1: “It’s better to go in your neck over your groin.” RESP5: “You can’t get water from old buildings because of [the] lead pipes.” RESP3: “Always inject towards your heart.” RESP4: “Blow on the pin when [you are] putting it in.”

In what ways could treatment be improved to ensure that those with needle fixation can effectively be treated for their addiction? In response to this question, a number of themes and subthemes were found.

Dislike of methadone When talking about treatment for heroin addiction, all respondents spoke about a dislike of methadone. These findings were responsible for the allocation of three sub-themes.

295

THE PUBLIC SPHERE | 2017 ISSUE

Stigma and inconvenience of consuming on site at a chemist Participants described the stigmatisation of having to wait to collect and consume their prescriptions in a busy chemist. RESP2: “This chemist I used to go to, they made me wait until all the other customers had gone before they gave me my methadone. Even though I collect it every day and it’s ready to pick up.” RESP3: “The other day I asked to go into the private room and the chemist just said ‘no they’re too busy’. So I had to drink it in front of everyone.” Three participants also talked about the inconvenience of having to go to the chemist each day to collect their prescription. RESP1: “It’s such a pain having to go every day, like I haven’t got anything else to do.” RESP5: “It’s fucking stupid right? They let me pick up for Sunday on Saturday, and if it’s a bank holiday I get Monday’s as well. And if it’s Christmas when everyone is out drinking and partying I can pick up like 4 or 5 days’ worth”.

Damage to teeth Some participants spoke about the negative consequences that taking methadone has on teeth. RESP1: “The worst thing about methadone – apart from getting addicted to it – is how bad my teeth have got from it.” RESP2: “My dentist said to me that it was the methadone doing it. I’ve had three teeth taken out this year.”

296

[Flushing, rituals and needle fixation among heroin addicts]

RESP5: “I’ve been on methadone for 15 years and… I’ve lost every one of my teeth apart from two because of it.”

Perceive the process as controlling Some respondents also specifically described methadone maintenance treatment as controlling. RESP5: “[The drug treatment centres] are the worst. The things they make you do for your script… They’ve made me sit in that waiting room for hours on a rattle with no money with people coming in selling it left right and centre.” RESP3: “They know they’ve got you by the bollocks, so they can treat you like shit.” RESP2: “The amount of times they cancelled appointments on me and then just expect that I can get in the next day no problem.” RESP3: “They used to let me get my prescription through the doctors [local GP], based on the fact that I had a relationship with the doctor that had built up over many years. Now they making me and everyone else come here where I see three different people.” RESP5: “I’ve been going to [drug treatment] services for 10 years and get the same dose every day and have to take it on site [at the chemist]… the service has changed… Now they make me come in once a fortnight and have told me I can’t take my methadone home ‘cause I’ve got to be clean first. I used to be trusted with a week’s worth. Now I can only get one day.”

297

THE PUBLIC SPHERE | 2017 ISSUE

Injectable methadone When considering solutions and other ways to treat opiate addiction, two themes were found. Some respondents suggested injectable methadone as a treatment solution. RESP5: “When I first started I was on injectable meth and I wish I still was now.” RESP4: “I prefer digging [injecting] stuff so injectable methadone would be best for me.” RESP1: “I can’t really think of anything that would help. [Pause]. Actually when they first started giving meth I think it was injectable. I think that’d be alright then at least you’d get a knock off it.”

Residential rehabilitation Another common theme suggested to treat opiate addiction was through residential rehabilitation. RESP1: “The only way to do it is to go to rehab.” RESP5: “I don’t think I’ve seen anyone ever stop taking meth. If you wanna get clean you’ve got to go to rehab.” RESP3: “I’d like to go to rehab to get clean but to be honest I’m not ready yet.”

298

[Flushing, rituals and needle fixation among heroin addicts]

DISCUSSION Why do some heroin users develop an addiction to the injecting process? Three themes were found in answer to why some people develop an addiction to the injecting process: previous obsessive traits, irrational superstitions, and insecure attachments. Previous research by Pates and Gray (2009) found that flushing the syringe and having injecting rituals are a precursor to having needle fixation; this study concurs with those findings. All participants in this study had some level of needle fixation and all had a very particular injecting ritual. Most also regularly flushed their syringes multiple times, believing that this improved the hit. Pates and Grayâ&#x20AC;&#x2122;s research shows that certain types of people may be more susceptible to developing this kind of behaviour. As this is a qualitative study that only interviewed people with needle fixation, it is impossible to say whether there is a causal link between obsessive or superstitious tendencies and needle fixation. However, my findings are in line with the theory that such preexisting tendencies may increase the likelihood of developing needle fixation. All the people interviewed had both needle fixation and irrational superstitions, and most demonstrated other obsessive characteristics. Many of the obsessive behaviours mentioned by the participants in this study were cleanliness-related. It could also be said that injecting in a ritualistic way is obsessive. Cleaning the home, hand washing, and tidying were all discussed by participants. As intravenous injecting is dangerous due in part to risks of infection and contamination, it could be the case that these rituals developed as a mechanism to ensure cleanliness during the injecting process, and then became more pervasive in the usersâ&#x20AC;&#x2122; everyday lives. There is a common conception that heroin

299

THE PUBLIC SPHERE | 2017 ISSUE

addicts are unclean, but this notion was challenged within the findings of this study. Interviews also probed superstitious behaviours, as these are often irrational and can affect people’s daily lives in a similar way to obsessive traits; they are carried out more as a matter of necessity rather than choice. Most participants in this study had many superstitions; the limited data herein suggests that people who have obsessive traits and are naturally superstitious may have an increased likelihood of developing needle fixation. Another theme discussed within the confines of this research relates to attachments. As mentioned in the literature review, some people with insecure attachments as children develop substance abuse problems in adulthood. This study furthers the theory, finding that three of the five interviewed had insecure attachments. It may be the case then, that not only do attachment issues increase the chance of substance misuse problems but also lead to an increased chance of developing needle fixation. More thorough and systematic study is warranted in this regard.

How do some people develop practices of rituals, flushing and needle fixation? To identify how these rituals take shape, the research focused on five themes that were discussed by the research informants: (i) incremental adaptions to injection rituals, (ii) ever increasing rituals, (iii) advice from friends, (iv) contradictory advice from professionals, and (v) the belief that flushing improves the hit. By asking participants to describe their exact injecting routine, the researchers attempted to break down each user’s practices to discover the deeper significance of each step. Findings were significant in part because no previous research had attempted to interrogate how these practices develop. The quotes

300

[Flushing, rituals and needle fixation among heroin addicts]

cited above highlight some interesting excerpts from the interviews that attempt to describe the main components of some rituals. Ultimately, each of the study participants noted incremental changes to their injection process over time. Similarly, this study found that for the heroin users involved, rituals are always expanding. When asked to describe the development of their routine, no participants mentioned that they had reduced their injecting ritual in any way. This is concerning because, as more extreme rituals become engrained in everyday practice, injection use becomes more difficult to roll back due to beliefs that it possesses a beneficial function. Most participants believed repeatedly flushing the syringe after the injection improved the hit and thus this became part of their routine. If a person truly believes this to be the case, then it would be difficult for them to stop. All participants flushed because they believed it improved the hit, although this is not the case (Preston et al. 2009). Some participants explained their routine by stating that it was based upon advice from addiction service and medical professionals. After further discussion, it was determined that professional advice is sometimes contradictory, with underlying effects upon a userâ&#x20AC;&#x2122;s practices. Alcohol wipes represent a recent example of changing advice for one participant; once provided to sterilise the skin before and after the injection, their distribution has since been withdrawn and their use is no longer encouraged. This contradictory advice is possibly due to the plethora of organisations that are involved in drug treatment and the fact that professional understanding of addiction and public health changes over time. To solve this problem, one agency should administer drug treatment. It would seem to be that there is so much contradictory advice from professionals and friends that drug users do not know how best to safely perform injections. Rituals develop from a combination of practices that have been recommended

301

THE PUBLIC SPHERE | 2017 ISSUE

to individuals by friends and professionals. An amalgamation of conflicting professional perspectives and changes informed by time saving, logic and convenience affect a user’s rituals. As a result, beliefs surrounding the injection practice remain diverse and often inconsistent across individuals. Ultimately there is significant variation in what people think is necessary, what constitutes good practice, what constitutes bad practice, and those customs that should be avoided. This absence of a coherent message delivered to users seeking treatment mirrors the wider state of drug treatment in the UK. There is no single agency in charge of collating the national data or standardising treatment. Furthermore, numerous organisations are involved in drug treatment: Public Health England, the Department of Health, the National Treatment Agency and the National Drug Treatment Monitoring Service. Lack of coordination seems to be an endemic problem of the drug treatment system in the UK. There is no overarching national strategy and no group accountability for failings.

In what ways could treatment be improved to ensure that those with needle fixation can effectively be treated for their addiction? In considering opiate treatment, this study uncovers a number of themes. There is a general dislike of methadone, which must invite designers of addiction treatments to think about an array of sub-themes: stigma and inconvenience of having to consume oral methadone at a chemist’s, damage to teeth, and the perception that methadone maintenance treatment is controlling. Other findings recommend the improvement of treatment through the reintroduction of injectable methadone and an increase in residential rehabilitation places. No existing research has documented drug users’ opinions on methadone. This study finds no positive comments in its regard. Participants talked about the stigma of having to go to a

302

[Flushing, rituals and needle fixation among heroin addicts]

pharmacist every day and stand in a busy shop while drinking methadone in front of other customers. According to qualitative accounts, this process is gravely demeaning and an inherently stigmatising practice. In one case, even though there was a free private room, the participant described the embarrassment of having to consume their methadone in the shop, as the pharmacist was too busy to see them privately. Another negative observation made about methadone is the damage it can inflict upon the userâ&#x20AC;&#x2122;s teeth. Methadone has a high sugar content and is known to cause tooth decay, although there are no current academic studies that identify a casual link (Ma et al. 2012). As some users are prescribed over 150mls of methadone every day, it is anticipated that some people will experience problems with their teeth (Reece 2007). In agreement with Lally (2003), this study also finds that MMT is perceived as a means of instituting control over users, rather than helping them to become drug free. Many respondents spoke about their negative experiences with the drug treatment company that provided their methadone prescriptions. Informants complained about cancelled appointments and rearranged meetings, generally booked on very short notice and deemed mandatory in order for users to receive their prescription. Userâ&#x20AC;&#x2122;s complaints about MMT are frequently attributed to the inconvenience of having to consume methadone onsite. Users questioned the necessity of these protocols arguing that they can take home their prescriptions on the days that the chemist is closed. They see the requirement to report to the chemistâ&#x20AC;&#x2122;s every day as an arbitrary and patronising rule, which can clearly be loosened as it is during holidays. The daily burden of visiting the chemist to take oral methadone seems to simply become part of the routine of a heroin addict. For treatment to result in a breaking of addiction, this routine must be broken as well;

303

THE PUBLIC SPHERE | 2017 ISSUE

what is needed may be the creating of an identifiable critical moment for transition or a final end point to treatment. Improving opiate treatment should include injectable methadone and more residential rehabilitation spaces. Participants disliked oral methadone treatment, and enrolment in addiction treatment centres is critically low. Injectable methadone was to be expected as a point of discussion, as many users were aware that it had been prescribed in the past and it more closely mimics the experience of using heroin. Because of this closer mimicry of heroin, it seems rational that injectable methadone would be prescribed to injecting drug users, as it may attract more users towards treatment centres and professional care. It may also help to associate methadone with being a drug substitute, rather than a less desirable alternative to illicit drug use. However, injectable methadone would not help to challenge an addiction to the injecting process – needle fixation – which must also be broken in order to truly overcome the addiction. It is interesting that residential rehabilitation was so broadly suggested, as it is a particularly difficult process for addicts. Residential rehabilitation programs can be very painful due to the harsh effects of withdrawal, and take weeks or months to complete. Thus, the suggested adoption of residential treatment mechanisms by interviewees indicates that some of those interviewed were determined to achieve abstinence from heroin. Furthermore, such recommendations serve to illuminate the severe dislike of oral methadone among heroin users.

POLICY RECOMMENDATIONS The most common form of opiate treatment is oral methadone treatment. This study finds that the heroin users interviewed disliked this treatment method for a number of reasons.

304

[Flushing, rituals and needle fixation among heroin addicts]

Injectable methadone was first prescribed in the 1980s and there are a small number of people still receiving this treatment. It would seem that injectable methadone would be more satisfying for those with needle fixation, as it enables them to fulfil both cravings. Furthermore, if users are going to be injecting regardless of treatment type, it is better to supply individuals with a medically tested substance than street heroinâ&#x20AC;&#x201D;which poses many risks to the user. Although there are concerns from a harm minimisation perspective, the overarching aim is to reduce injection of unsafe substances. This study suggests that those with insecure attachments are more likely to develop needle fixation. To deal with such findings, policymakers could introduce an early assessment tool to discover which service users have insecure attachments to attempt to alleviate this concern before it becomes a problem. An increase in residential rehabilitation centres may also serve to improve treatment outcomes. Residential rehabilitation acts as a critical turning point and can enable drug users to completely change their lifestyle, in contrast to initiatives that promote daily methadone consumption (Smyth et al. 2012). While this time and labour intensive intervention may prove costlier when considering monetary implications, increased expenditures associated with residential rehabilitation will lead to improved results and decreased costs in the long term. To streamline professional treatment schemes and policy recommendations, I suggest the creation of one central treatment agency to replace regional bodies operating across the UK, to govern policy pertaining to intravenous drug use. This step would enable the dissemination of a clear message directed at drug users and would help to combat current confused or counterproductive information in circulation.

305

THE PUBLIC SPHERE | 2017 ISSUE

CONCLUSIONS This study has generated new, valuable information on the subject of needle fixation. Previous research indicated that needle fixation can pose a sizable challenge for many heroin addicts, but there is no existing research that discusses how some addicts develop fixation. This study finds that some people develop needle fixation because of certain personality traits. Those with obsessive personality traits, superstitions, and insecure attachments are more susceptible to developing needle fixation. All participants in this study possess obsessive traits and were superstitious in their daily life. There was also found to be a possible link with insecure attachments as a child. These potential determinants are important to consider as their analysis could help experts identify individuals at risk for needle fixation. The way in which rituals developed was incremental and seemed to expand over time. Aspects of peoples’ routines were based on contradictory advice offered by professionals, advice from friends, and the view that flushing the syringe improves the hit. The risks associated with ever-expanding rituals can be minimised by ensuring that all advice given to drug users is the same by having one central agency to deal with drug treatment. As it stands, there are no prevalence estimates on the number of people who may have needle fixation. Evaluating the severity of this phenomenon would be an interesting area of further study. Another avenue for further study would be to conduct a participant observation study of people with needle fixation injecting to really find out about their routines and how they developed.

306

[Flushing, rituals and needle fixation among heroin addicts]

REFERENCES Advisory Council on the Misuse of Drugs (1982) Treatment and Rehabilitation. London: Her Majesty’s Stationery Office. Advisory Council on the Misuse of Drugs (1988) AIDS and drug use, part 1. HMSO: London. Al-Adwani A (2001) The Myth of Needle Fixation, Addiction 96: 335. Allen J, and Land D (1999) Attachment in Adolescence. In: Cassidy J, Shaver P (eds) Handbook of Attachment. New York: Guilford, (319-335). Attride-Stirling J (2001) Thematic Networks: an Analytic Tool for Qualitative Research. Qualitative Research 1(3): 385-405. Bell J (1999) Doing your research project: a guide for first-time researchers in education and social science. Open University Press: Buckingham. Benninghoff F, Solai S, Huissoud T, and Dubois-Arber F (2003) Evaluation de Quai 9 ‘Espace d’acceuil et d’injection’ à Genéve: période 12/2001–12/2000. Lausanne: Institut universitaire de médecine sociale et préventive. [Online]. Available at: http://www. premiereligne.ch/wp-content/uploads/2013/11/Quai9GE.pdf [accessed 23rd August 2016]. Borhani Y (2013) Substance Abuse and Insecure Attachment Styles: A Relational Study. Journal of Transdisciplinary Writing and Research from Claremont Graduate University 2(1): 1-13. Bowlby J (1969) Attachment and loss, Vol. 1: Attachment. New York: Basic Books. Braun V and Clarke V (2006) Using thematic analysis in psychology. Qualitative Research in Psychology 3(2): 77-101. Bryman A (2012) Social Research Methods. Oxford: Oxford University Press. Caspers K, Cadoret R, Langbehn D, Yucuis R, and Troutman B (2005) Contributions of attachment style and perceived social support to lifetime use of illicit substances. Addictive Behaviours 30(5): 1007-1011. Center for Substance Abuse Treatment (2005) Substance Abuse Treatment for Adults in the Criminal Justice System. Rockville: Maryland, USA. Connock M, Juarez-Garcia A, Jowett S, Frew E, Liu Z, Taylor RJ, FrySmith A, Day E, Lintzeris N, Roberts T, Burls A, and Taylor, RS (2007) Methadone and buprenorphine for the management of opioid dependence: A systematic review and economic evaluation. Health Technology Assessment 11(9): 1-171. Department of Health (2007) Drug misuse and dependence. UK guidelines on clinical management. Department of Health, UK. [Online]. Available at: http://www.nta.nhs. uk/uploads/clinical_guidelines_2007.pdf [accessed 23rd August 2016]. Dixon-Swift V, James EL, and Liamputtong P (2008) Undertaking Sensitive Research in the Health and Social Sciences. Cambridge: Cambridge University Press. DrugScope (2009) Drug treatment at the crossroads. DrugScope, UK. [Online]. Available at: http://www.drugscope.org.uk/Resources/Drugscope/Documents/PDF/Good%20 Practice/Drugtreatmentatthecrossroads.pdf [accessed 23rd August 2016]. Eley S, Beaton K, and McGiver G (2005) Co-operation in Drug Treatment Services: Views of Offenders on Court Orders in Scotland. The Howard Journal of Criminal Justice 44(4): 400-410. Farrimond H (2013) Doing Ethical Research. Basingstoke: Palgrave Macmillan. Field T (1996) Attachment and separation in young children. Annual Review of Psychology 47: 541-562.

307

THE PUBLIC SPHERE | 2017 ISSUE

Flick U (2009) An Introduction to Qualitative Research. London: Sage. Fraser S, Hopwood M, Treloar C, and Brener L (2004) Needle Fictions: Medical Constructions of Needle Fixation and the Injecting Drug User. Addiction Research and Theory 12(1): 67-76. Gossop M (2003) Drug Addiction and its Treatment. Oxford: Oxford University Press. Grossmann KE, and Grossmann K (1991) Attachment Quality as an Organizer of Emotional and Behavioral Responses in a Longitudinal Perspective. In: Parkes, CM, Stevenson-Hinde J, and Marris P (eds) Attachment Across the Life Cycle. London: Tavistock, 93- 114. Hammersley M (2003) Conversation analysis and discourse analysis: Methods or paradigms? Discourse & Society 14(6): 751-781. Hampl K (2002) Lecba zavislosti na heroinu metadonem v zarizeni AT bez luzkoveho zazemi. Ces.a Slov. Psychiat 98(2): 86-91. Hartnoll R, Mitcheson M, Battershy A (1980) Evaluation of heroin maintenance in controlled trials. Archives of General Psychiatry 37: 877-884. Hinton S, Signal T, and Ghea-White V (2013) Addicted to the needle: the relationship between needle fixation and impulsivity. Journal of Substance Use 18(1): 21-30. Hinton S (2015) Needle Fixation Profile: An Exploratory Assessment of Applicability in the Australian Context. Substance Use and Misuse 50: 1449-1452. HM Government (2010) Drug Strategy 2010: reducing demand, restricting supply, building recovery: supporting people to live a drug free life. [Online]. Drugs Strategy 2010, UK. Available at: https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/118336/drug-strategy-2010.pdf [accessed 23rd August 2016]. Hogg M, and Vaughan G (2008) Social Psychology. Essex: Pearson Education Limited. Home Office (2012) Putting Full Recovery First. [Online]. Available at: https://www.gov.uk/ government/uploads/system/uploads/attachment_data/file/98010/recovery-roadmap.pdf [accessed 23rd August 2016]. JRF (2006) ‘The Report of the Independent Working Group on Drug Consumption Rooms’. JRF, UK. [Online]. Available at: https://www.jrf.org.uk/sites/default/files/jrf/migrated/files/9781859354711.pdf [accessed 23rd August 2016]. Jupp V (2006) The Sage Dictionary of Social Research Methods. London: Sage Publications Ltd. Kassel J, Wardle M, and Roberts J (2006) Adult Attachment Security and College Substance Use. Addictive Behaviors 32(6): 1164-1176. Khoury L, Tang YL, Bradley B, Cubells JF, and Ressler KJ (2010) Substance use, childhood traumatic experience, and Posttraumatic Stress Disorder in an urban civilian population. Depression and Anxiety 27(12): 1077–1086. Krueger RA, and Casey MA (2000) Focus Groups. A Practical Guide for Applied Research. Thousand Oaks: Sage. Lally D (2013) ‘Long-term methadone use is form of ‘State-sponsored social control’. The Irish Times. [Online]. 21 October 2013. Available at: http://www.irishtimes.com/news/ ireland/irish-news/long-term-methadone-use-is-form-of-state-sponsored-socialcontrol-1.1567566 [accessed 23rd August 2016]. Larney S (2010) Does opioid substitution treatment in prisons reduce injecting-related HIV risk behaviours? A systematic review. Addiction 105(2): 216-223. Laurence J (2009) ‘The Big Question: Is methadone being over-prescribed as a treatment for drug addiction?’. The Independent, UK. [Online]. Available at: http://www. independent.co.uk/life-style/health-and-families/health-news/the-big-question-

308

[Flushing, rituals and needle fixation among heroin addicts]

is-methadone-being-over-prescribed-as-a-treatment-for-drug-addiction-1837156. html [accessed 23rd August 2016]. Lee E (2003) The Attachment System Throughout the Life Course: Review and Criticisms of Attachment Theory. Rochester Institute of Technology, USA. [Online]. Available at: http://www.personalityresearch.org/papers/lee.html [accessed 23rd August 2016]. Luck PA, Elifson KW, and Sterk CE (2004) Female drug users and the welfare system: a qualitative exploration. Drugs: Education, Prevention and Policy 11: 113-128. Ma H, Shi XC, Hu DY, Li X (2012) The poor oral health status of former heroin users treated with methadone in a Chinese city. Medical Science Monitor 14(4): 51-55. Marissen MA, Franken IH, Blanken P, van den Brink W, and Hendriks VM (2007) Cue exposure therapy for the treatment of opiate addiction: results of a randomized controlled clinical trial. Psychotherapy and Psychosomatics 76(2): 97-105. Marshall G (1994) Concise Oxford Dictionary of Sociology. Oxford: Oxford University Press. McBride A, and Pates R (2005) Needle Foucation: Deux ou Trois Choses que je Sais de Pica Manie (with apologies to Jean Luc Godard). Addiction Research and Theory 13(4): 395-402. McBride A, Pates R, Arnold H, and Ball N (2001) Needle fixation, the drug userâ&#x20AC;&#x2122;s perspective: A Qualitative Study. Addiction 96: 1049â&#x20AC;&#x201C;1058. McCann E, and Temenos C (2015) Mobilizing drug consumption rooms: inter-place networks and harm reduction drug policy. Health & Place 31: 216-223. Miles M and Hubermans A (1994) Qualitative Data Analysis: An Expanded Sourcebook. California: Sage. MSIC Evaluation Committee (2003) Final report of the evaluation of the Sydney Medically Supervised Injecting Centre. [Online]. Available at: http://www.druginfo.nsw.gov. au/__data/page/255/msic.pdf [accessed 23rd August 2016]. NDTMS (2015) Adult substance misuse statistics from the National Drug Treatment Monitoring System (NDTMS): 1 April 2014 to 31 March 2015. Public Health England, UK. [Online]. Available at: https://www.ndtms.net/Publications/downloads/Adult%20 Substance%20Misuse/adult-statistics-from-the-national-drug-treatment-monitoring-system-2014-2015.pdf [accessed 23rd August 2016]. National Drug Treatment Monitoring System (2016) Adult Primary Drug Clients: Treatment completion and no representation (% opiate clients). NDTMS, UK. [Online]. Available at: https://www.ndtms.net/default.aspx [accessed 23rd August 2016]. NHS (2003) Injectable heroin (and injectable methadone). Potential roles in drug treatment. NHS, UK. [Online]. Available at: http://www.nta.nhs.uk/uploads/nta_injectable_heroin_and_methadone_2003_fullguide.pdf [accessed 23rd August 2016]. NICE (2007) Methadone and buprenorphine for the management of opioid dependence. [Online]. National Institute for Health and Clinical Excellence, UK. Available at: http://www.nice.org.uk/guidance/ta114/resources/guidance-methadone-and-buprenorphine-for-the-management-of-opioid-dependence-pdf [accessed 23rd August 2016]. NTA (2007) Drug misuse and dependence: UK guidelines on clinical management, National Treatment Agency, UK. [Online]. Available at: http://www.nta.nhs.uk/uploads/ clinical_guidelines_2007.pdf [accessed 23rd August 2016]. National Treatment Agency (2012) The Role of Residential Rehab in an Integrated System. [Online]. Available at: http://www.nta.nhs.uk/uploads/roleofresi-rehab.pdf [accessed 23rd August 2016]. Pates R, and Gray N (2009) The Development of a Psychological Theory of Needle Fixation. Journal of Substance Misuse 14(5): 312-324.

309

THE PUBLIC SPHERE | 2017 ISSUE

Pates R, Arnold K, and McBride A (2009) The Identification of Needle Fixation: The Development of the NEFPRO, a Clinical Screening Tool. Journal of Substance Misuse 14(5): 306-311. Pates R, McBride AJ, Ball N, and Arnold K (2001) Towards an Holistic Understanding of Injecting Drug Use: An Overview of Needle Fixation. Addiction Research and Theory 9(1): 3-17. Pates R, McBride A, and Arnold K (2005) Injecting Illicit Drugs. Oxford: Blackwell Publishing Ltd. Patton M (1990) Qualitative evaluation and research methods. Thousand Oaks, CA, US: Sage Publications Incorporated. PHE (2013) Drug Statistics from the National Drug Treatment Monitoring System (NDTMS) 1 April 2012 to 31 March 2013. National Treatment Agency, UK. [Online]. Available at: www.nta.nhs.uk/uploads/annualdrugstatistics2012-13-statisticalreport. pdf [accessed 23rd August 2016]. Preston A, Derricott, J, and Hardarce, P (2009) Safer Injecting Reducing Your Risk. National Treatment Agency, UK. [Online]. Available at: http://www.nta.nhs.uk/uploads/ hrdvd5.pdf [accessed 6th December 2016]. Puigdollers E, Domingo-Salvany A, Brugal M, Torrens M, Alvaros J, Castillo C, Magri N, Martin S, and Vazquez J (2004) Characteristics of heroin addicts entering methadone maintenance treatment: Quality of life and gender. Substance Use and Misuse 39: 1353-1368. Reece AS (2007) Dentition of addiction in Queensland: poor dental status and major contributing drugs. Australian Dental Journal 52(2): 144-149. Robson C (2011) Real World Research. Chichester: Wiley. Roche C (2014) Reducing Harm: The case for drug consumption rooms in the UK. Talking Drugs, UK. [Online]. Available at: http://www.talkingdrugs.org/youth-rise/reducing-harm-the-case-for-drug-consumption-rooms-in-the-uk [accessed 23rd August 2016]. Seale C (2004) Social Research Methods: A Reader. Oxon: Routledge. Schatz E, and Nougier E (2012) Drug consumption rooms Evidence and practice. IDPC Briefing Paper. International Drug Policy Consortium. [Online]. Available at: http:// www.drugsandalcohol.ie/17898/1/IDPC-Briefing-Paper_Drug-consumption-rooms. pdf [accessed 23rd August 2016]. Silverman D (2000) Doing Qualitative Research. London: Sage Publications Ltd. Stewart T (1987) The Heroin Users. London: Pandora Press. Strang J, Metrebian N, Lintzeris N, Potts L, Carnwath T, Mayet S, Williams H, Zador D, Evers R, Groshkova T, Charles V, Martin A, and Forzisi L (2010) Supervised injectable heroin or injectable methadone versus optimised oral methadone as treatment for chronic heroin addicts in England after persistent failure in orthodox treatment (RIOTT): a randomised trial. The Lancet 375(9729): 1885–1895. Smyth B, Fagan J, and Kernan K (2012) Outcome of heroin-dependent adolescents presenting for opiate substitution treatment. Journal of Substance Abuse Treatment 42(1): 35-44. Thorberg F and Lyvers M (2006) Attachment, Fear of Intimacy and Differentiation of Self Among Clients in Substance Disorder Treatment Facilities. Addictive Behaviours 31(4): 732-737. Treffurth Y, and Raj Pal H (2010) A Case of Needle Fixation. British Medical Journal Case Reports 10: 1136.

310

[Flushing, rituals and needle fixation among heroin addicts]

Tronick E, Ivey P, and Morelli G (1992) The Efe Forager Infant and Toddlerâ&#x20AC;&#x2122;s Pattern of Social Relationships: Multiple and Simultaneous. Developmental Psychology 28(4): 568577. UKDPC (2008) Reducing Drug Use, Reducing Reoffending: Are programmes for problem drug-using offenders in the UK supported by the evidence? [Online]. Available at: http://www.ukdpc.org.uk/wpcontent/uploads/Policy%20report%20%20Reducing%20drug%20use,%20reducing%20reoffending%20(summary).pdf [accessed 23rd August 2016]. UKDPC (2010) Getting Serious about Stigma: the problem with stigmatising drug users. An Overview. UK Drug Policy Commission, UK. [Online]. Available at: http://www.ukdpc.org.uk/wp-content/uploads/Policy%20report%20%20Getting%20serious%20 about%20stigma_%20the%20problem%20with%20stigmatising%20drug%20 users.pdf [accessed 23rd August 2016]. Van Ijzendoorn M, and Kroonenberg P (1988) Cross Cultural Patterns of Attachment: A Meta-Analysis of the Strange Situation. Child Development 59: 147-156. Warburton H, Turnbull PJ, and Hough M (2005) Occasional and Controlled Heroin Use: Not a Problem? York: Joseph Rowntree Foundation. Ward J, Hall W, and Mattick R (1999) Role of maintenance treatment in opioid dependence. The Lancet 353(9148): 221-226. Weber RP (1985) Basic Content Analysis. New Delhi: Sage. Wengraf T (2001) Qualitative Research Interviewing: Semi-Structured: Biographical and Narrative Methods. London: Sage. Winsome St.J, and Johnson P (2000) The Pros and Cons of Data Analysis Software for Qualitative Research. Journal of Nursing Scholarship 32(04): 393-397.

311